Abstract
The United States Food and Drug Administration's assessment strategy for the determination of the carcinogenic potential of drug residues in food‐producing animals is embodied in its Human Food Safety Policy. This policy calls for the utilization of a threshold‐assessment procedure to determine whether a veterinary drug possesses carcinogenic potential. Chronic lifetime studies in rodents must be performed if it is deemed that the agent may possess carcinogenic properties. A virtually safe level (VSL) of exposure for man is established by applying a modified statistical linear‐extrapolation model to the carcinogenic dose‐response relationship seen in the test‐animal studies. Identification of the major metabolites in the total drug residue in the target animals is required, and the same process applied to the parent drug may be applied to the metabolites. If a carcinogenic metabolite is found in the edible tissue which is significantly more potent and/or more persistent than the parent drug, the VSL is established for the metabolite rather than for the parent compound. A rigorous identification of metabolites is required, which is followed by the development of analytical methodology to ensure that carcinogenic residues in tissue are below the VSL.