Using Pharmacokinetics to Assure Chemical Food Safety

Using Pharmacokinetics to Assure Chemical Food Safety

As a veterinarian, I’m responsible for the health and welfare of my animal patients. Sometimes, drugs are used to treat animals that are being raised for food (e.g., meat and milk). Therefore, tissue residues are a unique concern in veterinary medicine because indirect exposure to drugs and their metabolites through eating meat or milk could potentially negatively impact human health. For this reason, we need to be able to make robust predictions of the time delay for tissue drug residues to fall below concentrations that have been shown to be safe for human consumption (the “tolerance”). The plasma elimination half-life is the pharmacokinetic (PK) parameter that reflects a drug’s persistence in the body. A challenge for predicting tissue residues is that they may persist beyond when plasma concentrations can be detected with even the most sensitive bioanalytical method.

Compartmental PK models can be used to describe plasma time-concentration data. These models capture the different rates at which a drug distributes to and from the various tissues of the body. These rates are dependent on how the physico-chemical properties of the drug interact with the characteristics of each tissue. In compartmental PK models, a compartment represents a group of tissues to which the drug distributes and equilibrates at the same rate. The number of compartments in the model determines the number of exponential terms needed to describe the plasma time-concentration curve.

As analytical techniques become more sensitive, ever lower plasma drug concentrations can be measured, and more compartments may be needed to fully describe a drug’s plasma time-concentration profile. If the analytical method is sensitive enough, it becomes possible for the terminal elimination phase to represent the half-life and persistence of the drug in the tissue from which it depletes the slowest (deep compartment), making it relevant to human food safety. Examples of groups of drugs for which the number of compartments increased with increasing sensitivity of the bioanalytical technique include the antibacterial tetracyclines and aminoglycosides1,2.

Using a published model for oxytetracycline1, the consequences of using a two- versus a three-compartment pharmacokinetic model to predict tissue drug concentrations is illustrated in Figures 1 and 2, respectively. Notice that the two-compartment model greatly under-predicts the tissue concentrations and the time needed for them to deplete to levels that are below the tolerance. For this reason, regulatory withdrawal times (official withdrawal times that appear on the label of a pharmaceutical product approved for food-producing animals) must be based on tissue data collected from animals that are sacrificed at sequential times after treatment. Models based on plasma data can be used to estimate an appropriate waiting time before animals can be slaughtered following the extra label use of a drug, and they have the advantage that animals do not need to be sacrificed. But this is only possible if the analytical method is sensitive enough to pick up the concentrations associated with the deepest compartment that represents the tissue from which the drug depletes the slowest.


[1] Meijer LA, Ceyssens KG, de Jong WT, de Grève BI. Three phase elimination of oxytetracycline in veal calves; the presence of an extended terminal elimination phase. J Vet Pharmacol Ther. 1993 Jun;16(2):214-22. PubMed PMID: 8345571.

[2] Wenk M, Spring P, Vozeh S, Follath F. Multicompartment pharmacokinetics of netilmicin. Eur J Clin Pharmacol. 1979 Nov;16(5):331-4. PubMed PMID: 520400.

To learn more about using allometric scaling to determine first-in-man dosing data from animal PK data, please read this case study.

Ronette Gehring

About the Author

Ronette Gehring

More Posts by This Author

Ronette Gehring is an associate professor of clinical pharmacology at the Kansas State University College of Veterinary Medicine, where she directs the Midwest Regional Center of the Food Animal Residue Avoidance Databank (FARAD). Her research interests lie with using computer-based modeling as a quantitative framework that integrates and explains pharmacokinetic and pharmacodynamic data based on current scientific understanding in veterinary and comparative pharmacology.