Presentation Abstract

An Informatic Approach to Estimating Ecological Risks Posed by Pharmaceutical Use
Mitchell Kostich, James Lazorchak, and Gregory Toth
National Exposure Research Laboratory, Office of Research and Development,
U.S. Environmental Protection Agency,
Cincinnati, OH

Pharmaceuticals administered to humans and other animals are often excreted from treated organisms as intact drug or as active metabolites. Some of these active materials have been shown to make their way into the environment. The environmental concentrations of the vast majority of pharmaceuticals and their metabolites, however, are not known. The sensitivity of native organisms to chronic exposure to the doses of active material likely to be found in the environment also is typically not known. Direct determination of these important parameters is too expensive to perform on the entire pharmacopoeia, and a rational way of prioritizing individual drugs for more detailed study is needed.

For many drugs, sufficient information is available in the scientific literature and within regulatory filings to estimate the upper bound of environmental concentrations and most probable mechanisms of environmental toxicity. Although these estimates involve large confidence intervals, they are quickly and cheaply produced and are probably the best available criteria for prioritizing drugs for more expensive direct tests of environmental impact.

This research project describes a new method similar to that employed by the European Union and by the U.S. Food and Drug Administration for estimating risks of human prescription pharmaceuticals based on information found in regulatory filings, as well as scientific and trade literature. Available data on usage patterns, metabolic transformation, and physical/chemical properties are fed into models of dilution, degradation, partition between matrices, and bioavailability to estimate effective environmental concentrations. Available data on mechanisms of action and modes of toxicity are considered together with cursory phylogenetic analysis to estimate the sensitivity of select native organisms to the estimated highest likely environmental concentrations of each drug.

Scores produced by this procedure will be used to prioritize pharmaceuticals for more detailed analytical and toxicological followup. The potential application of modified versions of this method to over-the-counter pharmaceuticals and veterinary pharmaceuticals also is described.

Top of page