How to extrapolate a withdrawal time from an EHSLC published detection time: a Monte Carlo simulation appraisal

For legitimate medications, veterinarians must advise the owners or trainers of horses on appropriate withholding times after a treatment, to avoid the risk of incurring a positive drug test.

To explore the safety span to select that a veterinarian may extrapolate a tailored withdrawal time (WT) from a generic detection time (DT) as published by the European Horserace Scientific Liaison Committee (EHSLC).

Using Monte Carlo simulations, it was shown that for a low variability of pharmacokinetic parameters (CV=20%), an uncertainty span of about 40% may be selected to transform a mean DT into a WT (i.e. WT=1.4 DT), which covers 90% of the horse population. In contrast for a highly variable drug (CV=40%), an uncertainty factor of about 2.1-2.2 needs to be selected, i.e. a WT that is twice the DT.

The relative impact of the different factors of variability on the final WT was documented by a so-called sensitivity analysis. It was shown that the parameters that have the greatest influence on the value of a DT are those that control the terminal half-life of the drug disposition. In contrast, parameters controlling the level of urine (or plasma) concentrations (i.e. the actual administered dose, the urine-to-plasma ratio and the bioavailability) collectively have a minimal influence on the DT.

In practice, this means that the main sources of uncertainty are of biological origin and cannot be reduced by any managerial options. The influence of the number of experimental horses that are used by EHSLC to establish a DT was shown that with the standard EHLSC protocol of 6 horses, half of the trials lead to a proposed DT that is equal to or higher than the population 90(th) percentile. Increasing the number of investigated horses to 8 and 10 would increase this last probability to 85 and 90%, respectively.