Administration Studies in Equine Antidoping Research: Designing Scientific Investigations to Effectively Direct Medication Control in Racehorses

ABSTRACT

Pharmacokinetics is the study of the movement of drug in the body and includes the processes of absorption, distribution, metabolism, and excretion. Pharmacodynamics is the pharmacologic effect of the drug on the body. The pharmacokinetics of a drug determines its pharmacologic effect. Pharmacokinetic studies describe drug concentrations while pharmacodynamics allow for assessment of drug effects. Combined pharmacokinetic/pharmacodynamic studies allow for integration of drug concentrations with pharmacologic effect. Data generated from pharmacokinetic studies can be especially useful in establishing regulatory recommendations, determining appropriate thresholds, screening limits, administrative stand down times, and corresponding detection times. To generate the appropriate information, the following must be considered (1) the test subjects (i.e., number, age, breed, and fitness level), (2) selection of an appropriate dose/route and duration of administration, (3) sample matrix (blood, urine, and hair), (4) time(s) of sample collection, (5) development of an analytical method with appropriate sensitivity, and (6) what analytes to measure (parent and/or metabolite). Pharmacokinetic studies generate drug concentration data that can be used to calculate key pharmacokinetic variables necessary for establishing screening levels and detection times. Pharmacodynamic assessments can aid in understanding the pharmacologic effects of drugs and in correlating drug concentrations to these effects. Various models, including in vivo (whole animal), in vitro, and ex vivo assessments, can be utilized to determine pharmacodynamic effects. Factors to consider in the design of pharmacokinetic studies, basic pharmacokinetic parameters, and examples of pharmacodynamic assessments will be discussed in detail during this tutorial.

1. Introduction

Regulatory recommendations are based on the pharmacology of the drug, specifically the pharmacokinetics (PK) and to some extent the pharmacodynamics (PD) of the compound. Pharmacokinetics is the study of the movement of a drug in the body and includes the processes of absorption, distribution, metabolism, and excretion (ADME). Pharmacodynamics is the pharmacologic or physiologic effect of the drug and can include things such as sedation and alleviation of pain and inflammation in addition to adverse or “off target” effects. Pharmacokinetics and PD are interrelated processes in that it is the PK of a drug that determines its PD effect. The concentration of a drug that reaches a particular site (target and off‐target) is dependent on its ADME, and it is the concentration that ultimately determines the observed response (magnitude and duration).

Pharmacokinetic studies describe drug concentrations while pharmacodynamic studies allow for assessment of drug effects. Combined pharmacokinetic/pharmacodynamic (PK/PD) studies allow for integration of drug concentrations with the pharmacologic effect. Data generated from PK studies can be especially useful in establishing regulatory recommendations, determining appropriate thresholds, screening limits (SL), administrative stand down times, and corresponding detection times (DT). To generate the appropriate information, the following must be considered (1) the study subjects (i.e., number, age, breed, and fitness level), (2) selection of an appropriate dose/route and duration of drug administration, (3) sample matrix (blood, urine, and hair), (4) time(s) of sample collection, (5) development of an analytical method with appropriate sensitivity, and (6) determination of analytes to be measured (parent and/or metabolite). Drug concentration data generated from PK studies are used to calculate key pharmacokinetic parameters necessary for determination of SL and DT. Inclusion of PD assessments can aid in understanding the pharmacologic effects of drugs and in correlating drug concentrations to these effects. Various models, including in vivo (whole animal), in vitro, and ex vivo, assessments can be utilized in these studies.

Factors to consider in the design of pharmacokinetic studies, basic pharmacokinetic parameters, and examples of pharmacodynamic assessments are discussed below. Select examples of what can be learned from PK/PD studies are also briefly discussed.

2. Considerations in Pharmacokinetic/Pharmacodynamic Study Design

2.1. Horses

An important aspect in the design of PK/PD studies for establishing regulatory recommendations is horse selection. The sample size for administration studies can be highly variable, and because these studies are generally descriptive in nature, without a hypothesis, power calculations are not typical. Studies supported by the Racing Medication and Testing Consortium (RMTC) in the United States, typically use anywhere from 12 to 20 horses. Studies conducted in support of the European Horserace Scientific Liaison Committee (EHSLC) tend to use 6 as a minimum number. In general, the greater the number of horses utilized in an administration study, the greater the confidence that the generated values represent a greater portion of a given horse population. In other words, a larger sample size increases the predictability of the DT and therefore the withdrawal time (WDT) [1].

The age of the study subjects is an important consideration as age has the potential to effect drug clearance. Age‐related changes in hepatic metabolism and renal clearance can affect the residence time of a drug in the body. Flunixin meglumine is an example of a drug whereby total body clearance and the rate of elimination were significantly greater in a younger population of horses (≤ 5 years of age) compared with an older group (> 9 years of age) [2]. While the age range may vary between studies, historically, studies used for establishment of regulatory recommendations typically target an age range of 2–3 years on the low end and 6–7 years on the high end. Although not well studied in horses, breed may also affect the PK and/or PD of certain drugs, perhaps duet to genetic differences in metabolic enzymes, transporters, or target receptors.

2.2. Exercise and Fitness Status

Exercise and training status can lead to alterations in the pharmacokinetics of drugs. This has been well documented in human medicine with a lesser number of reports in horses. For example, the distribution of detomidine is increased and the rate of elimination decreased in horses receiving an intravenous dose immediately after exercise, compared with those receiving the same dose at rest [3]. Conversely, the pharmacokinetics of furosemide when administered to horses enrolled in a regular exercise program does not appear to differ when administered at 4‐ and 24‐h prior to a treadmill workout [4, 5]. In another study, training status (chronic exercise) does not appear to impact the elimination of flunixin meglumine [2].

Although the impact of exercise on drug pharmacokinetics is not clear‐cut and depends on the physiochemical characteristics of the drug being studied and exercise‐related conditions [6], it is an important consideration in the design of studies utilized to establish recommendations. For the purposes of this discussion, it is important to differentiate between “exercise” and “training status.” In this review, exercise will refer to a single event or exertion whereas training status refers to chronic exercise leading to more long‐term physiological adaptations that may affect drug pharmacokinetics when a drug is administered to the horse at rest.

In humans, exercise can affect drug distribution [6]. During a single exercise session, decreases in plasma volume have been reported [7], whereas plasma volume increases with chronic exercise [8]. In either situation, drug distribution may be affected. Exercise has been shown to influence hepatic blood flow in both humans and horses [6, 9, 10]. Using bromsulphalein (BSP) as a marker, investigators demonstrated a decrease in hepatic blood flow in horses following exercise along with a decrease in the volume of distribution of BSP at steady state [9]. In a subsequent study the investigators demonstrated a decrease in BSP clearance and subsequently an increase in the elimination half‐life as exercise intensity increased [10]. The same group of investigators assessed the pharmacokinetics of antipyrine in a group of sedentary horses and a group of horses that were in active training [11]. When administered at rest, the clearance and volume of distribution of antipyrine were significantly increased in the fit horses, compared with the nonexercising horses [11].

Exercise can also impact renal clearance and therefore the pharmacokinetics of a drug. This has been demonstrated in humans, whereby decrease in renal blood flow and glomerular filtration has been reported [12, 13]. Gleadhill and colleagues [14] reported decreases in the glomerular filtration rate (GFR) with mild exercise, presumably due to greater diversion of blood flow to muscles and skin. Depending on drug characteristics, urine pH can have a significant impact on renal clearance [15]. Changes in urinary pH is another is another well‐documented effect of exercise, both in humans and horses [16, 17, 18]. Increases in urinary pH have been shown to increase urine concentrations of phenylbutazone and its metabolites in horses [18]. In contrast to the above‐reported effects of exercise on renal clearance, long‐term exercise in humans appears to have little impact on the renal elimination of drugs when taken at rest.

Potential differences in ADME have led some investigators to utilize exercised research horses for administration studies in which the goal is to generate data that can be used for establishing regulations. Investigators incorporate works on a high‐speed treadmill to maintain fitness. While, admittedly, these horses may not be quite as fit as a horse that is actively racing, they are more representative than most sedentary research horses. Assessment of fitness levels can be accomplished by performing regular fitness assessments including measuring heart rate at speed and lactate concentrations at maximal exertion [19]. The V200, or the speed reached by a horse when its heart rate is 200 bpm, can be used to assess cardiovascular capacity. An increasing V200 indicates a higher level of fitness. Measuring serum lactate during and following maximum exertion is also a useful component of a fitness test. Typically, a “step test” or incremental velocity exercise test is employed, whereby exercise intensity is increased during a workout by increasing the velocity at set time points, and lactate concentrations are measured as the velocity increases. Blood lactate assessments allow for evaluation of aerobic capacity.

2.3. Selection of Dose, Route, Site, and Condition of Administration

An important aspect in planning PK/PD studies is selection of an appropriate dose, route of administration and whether single or multiple dose administration is most appropriate. Unfortunately, it is not possible to assess every possible dose or route of administration for a specific drug. Typically, if the drug is labeled for use in the horse, the manufacturers label dose, route of administration, and dosing interval (if drug is intended for multiple doses) are studied. In some instances, it may be necessary to use a drug in a manner other than that described by the manufacturer (extra‐label use) to treat a racehorse or perform diagnostic procedures. In this case, surveying racetrack practitioners about the most commonly used dose can be useful in determining the best administration protocol. Of equal importance to the dose and route of administration is the site and conditions of drug administration. For example, when administering a drug intramuscularly, the specific muscle group must be considered, as this can have a significant impact bioavailability. Firth and colleagues demonstrated this by administering procaine penicillin G at 4 different muscles. Bioavailability was 113% when administered in the neck (cervical serratus ventralis), 78% in the gluteal muscle, 97.6% in the biceps, and 94.2% in the pectoralis muscle [20]. Consideration of the conditions surrounding drug administration are also important. For example, digestive status (fed versus fasted or time of administration post feeding) can influence bioavailability following oral administration for some drugs but not for others. Meloxicam is an example of a drug whereby food does not affect absorption [21], whereas the absorption of flunixin meglumine or phenylbutazone is known to be impacted by the presence of food in the gastrointestinal tract [22, 23].

2.4. Sample Matrix

The most common biological matrices used for regulation of therapeutic and illicit substances are blood (serum or plasma) and urine. Urine can be advantageous because of the kidney's ability to concentrate compounds, making them present at higher concentrations compared with blood, thereby improving detection capabilities. However, it is important to note that the interpretation of urine concentrations is more difficult than plasma concentrations when regulating medications with the intent of guaranteeing the absence of a pharmacologic effect. There are a number of factors that confounding factors that should be taken into consideration including, among other things, the volume of urine which can be affected by hydration status and the pH of the urine. Blood is useful in that concentrations are more reflective of the pharmacologic effect. Although, notably, the sensitivity of analytical instrumentation continues to improve, drug concentrations in blood are generally lower than urine, which in some cases can make detection more challenging. Although often not logistically feasible because of the requirement to have known volumes, as opposed to just a “snapshot” of urine concentrations at a given time, one could simultaneously model urine and plasma data with a nonlinear mixed effects pharmacokinetic model (NLME). This would provide a Bayesian estimation of the slope of the terminal phase where the sensitivity of the analytical method precludes detection.

Knowledge of the matrix used for regulation of drug administration can aid in the selection of the most appropriate matrix for administration studies. In many cases, both blood and urine are collected. When collecting blood, selection of an appropriate collection tube is an important consideration [24, 25]. In the case of serum, it is important to remember that some drugs may bind to the silicone plug found in serum separator tubes, which can lead to artificially low concentrations on analysis. With plasma samples, it is important to consider the effect of the specific anticoagulant. Sample for detection of procaine penicillin, for instance, should be drawn into tubes containing sodium fluoride to inhibit plasma esterase activity [26, 27]. In another study, investigators demonstrated a significant difference in phenylbutazone concentrations when collected in tubes containing heparin versus Ethylenediaminetetraacetic acid (EDTA) [28]. When using Ca‐EDTA tubes, it is important to consider that some drugs may chelate with Ca, leading to artificially low concentrations. Another consideration is the method of blood collection. For ethical reasons, when collecting the large number of samples for pharmacokinetic analysis, an indwelling venous catheter is often utilized. While arguably this is more desirable than direct venipuncture, it is important to recognize that for some drugs, indwelling catheters may impact pharmacokinetics [29].

Hair, although not routinely used for drug testing, can be a good indicator of whether certain drugs have been administered and the approximate time of administration relative to sample collection. This might be for entry into a race or for out of competition testing or when a horse changes ownership.

2.5. Timepoints

Selection of time points for drug concentration determinations is largely dependent on the goal of the study. For example, if the goal of the study is to fully describe the PK profile, a larger number of samples is necessary, whereas if the goal is to only characterize the half‐time of the terminal phase, a smaller number of samples, collected during the elimination phase, may be adequate. The route of administration and formulation are also important considerations when selecting time points. If interested in the full PK profile and the drug is administered IV, where absorption is not a factor, samples are often collected earlier compared with administration via an extravascular route. However, to characterize the absorption phase most effectively, it is imperative to collect enough samples during this phase, so as not “miss” the maximum concentration (C_max) and time of maximum concentration (T_max) and to enable calculation an accurate rate of absorption (K_a). In the case of a drug in which the rate of elimination is more rapid than the rate of elimination (flip‐flop kinetics), it is important to remember that it is the slope of the terminal phase that is used to measure K_a. While this may vary somewhat depending on the drug, collecting a minimum of 3–4 time points per phase (absorption, distribution, and elimination) is a good “rule of thumb.” If the data is to be used for calculation of an irrelevant plasma concentration (IPC; see below), it is important to include enough time points to allow for an accurate calculation of the area under the curve (AUC) [30].

To characterize elimination most effectively, it is helpful to collect several samples in the terminal phase and quantitate concentrations down to the lower limit of quantitation. Ideally, if the half‐life is known, samples would be collected for 4–5 elimination half‐lives, necessitating minimal extrapolation of the terminal curve. Minimal extrapolation is also important when calculating the AUC, where the in general an acceptable extrapolation is between 15% and 25%. With some of the slower release formulations, it may be necessary to collect samples longer to obtain the entire elimination profile.

Pharmacokinetic studies involving multiple administrations of a drug may include additional time points beyond that for a single administration. Ideally, several timepoints after the first dose would be collected (as with a single administration) followed by peak (C_max) and trough (C_min; minimum concentration prior to next dose) concentrations up to the last dose and then multiple samples after the final dose. This allows for comparison of pharmacokinetic parameters after the first and last dose to determine if there is evidence of nonlinear kinetics that could influence drug exposure.

2.6. Analytical Method

A validated, sensitive analytical method utilizing certified reference standards is imperative to ensure the accuracy of the measured concentrations. For specific criteria, the reader is referred to other publications and industry specifications.

2.7. Pharmacokinetic Parameters and Regulatory Recommendations

With the appropriate study design, pharmacokinetic studies allow for determination of pharmacokinetic parameters. The C_max and T_max are two variables that can be determined from review of the concentration data following extravascular administration. The accuracy of these values is dependent on adequate characterization of the absorption phase of the concentration time curve. The AUC is a measure of total systemic exposure following drug administration. The AUC is dependent on bioavailability and overall drug clearance. Bioavailability represents the fraction of drug absorbed following extravascular administration and can be determined with the following equation:

$$\mathrm{F}=\frac{{\mathrm{AUC}}_{\text{extravascular}}}{{\mathrm{AUC}}_{\text{intravenous}}}\mathrm{x}\frac{{\text{Dose}}_{\text{intravenous}}}{{\text{Dose}}_{\text{extravascular}}}$$

The volume of distribution is a proportionality constant between a plasma concentration and the corresponding amount of blood in the body [31], and total systemic clearance is the ratio of the rate of drug elimination to the plasma concentration [32]. The half‐life of elimination is the time it takes for plasma or serum drug concentrations to decline by half in the terminal phase.

Pharmacokinetic parameters can be used to establish screening limits (SL). Using the Toutain approach [33], an IPC can be determined. Calculation of the IPC starts with determination of the effective plasma concentration (EPC). The EPC is calculated using the equation:

$$\mathrm{EPC}=\frac{\left(\text{Therapeutic Dose}\right)}{\mathrm{CL}\mathrm{x}\text{dosing interval})}$$

An alternative approach to calculate the EPC, when concentration data from intravenous administration is not available, is to use the equation:

$$\mathrm{EPC}=\frac{\mathrm{AUC}}{\text{Dosing Interval}}$$

Based on the Toutain approach, a safety factor of 500 is then applied to the EPC to determine the IPC:

$$\mathrm{IPC}=\frac{\mathrm{EPC}}{500}$$

Provided analytical instrument sensitivity allows, the IPC is often used as the SL. A DT corresponds to the time after drug administration where concentrations of drug in plasma or serum from all horses in an experimental study fall below the SL. This is a raw experimental observation. The DT is not the same thing as the withdrawal time WDT, which is the time following administration of a drug that a horse should not race to prevent a positive drug test. In general, the WDT is prolonged compared with the DT, as it is meant to take into consideration animal factors such as age, gender, breed, training, fitness, and so on. As DTs are based on a small number of horses, compared with a larger population of racehorses, it is highly likely that the SL will be exceeded by some horses, using published DT. For example, using Monte Carlo simulations, if the DT obtained from a 6‐animal study is used as the withdrawal time (WDT), there is a 50% chance that an animal in the 90th quantile of the population will exceed the SL [34]. As the number of animals in a study increases, the probability of exceeding the SL decreases.

Although recommendations are derived from data generated from very carefully designed PK studies, it is important to remember that they are only recommendations and can be very specific. Published DTs are specific to the route of administration, the dose administered, and the frequency of administration. The specific form or formulation may of a drug may also impact the regulatory recommendation, especially if a drug slow release. In the case of slow‐release formulations, in general, a drug can be detected for a longer period of time compared with immediate release formulations. Veterinarians should take into consideration the potential for nonlinear pharmacokinetics when recommending WDTs that are based on extrapolation from drug doses other than that for which there is a published detection time and/or concentration data. Nonlinear elimination can prolong the half‐life and therefore the detection time. Additionally, due to the inherent variability in products, veterinarians should exercise extreme caution if making recommendations for compounded products, based on DTs established for approved products (i.e., approved by the Food and Drug Administration). Even with approved formulations deemed bioequivalent, there is no guarantee that the same DTs or WTs are appropriate [35].

2.8. Pharmacodynamic Assessments

When the goal is to understand the effects of the drug on the body, a pharmacodynamic assessment may be included. Measures of pharmacodynamic effects can be objective or subjective in nature and may include whole body assessments or measurements of biomarkers, including changes in protein or genomic biomarkers. The type of assessments utilized depend on the drug, specifically what the intended therapeutic purpose is. In vivo assessments include things like lameness evaluations, effects on heart rate or assessments of the effects on noxious stimuli (e.g., thermal or mechanical nociception). Evaluating the effects of drugs on biomarkers, whether genomic or protein in nature, represents a more objective approach to assessing the pharmacodynamic effect of a drug. One example of a biomarker that has been used quite a bit in equine studies, specifically to assess effects of corticosteroids, is cortisol [36, 37]. The use of cortisol as a pharmacodynamic assessment is based on the phenomenon of suppression of endogenous cortisol following exogenous administration of corticosteroids and the fact that it takes time for endogenous production to return to normal following discontinuation of exogenous corticosteroids.

A second example of a biomarker assay that has been used in horses for assessing the effect of anti‐inflammatory drugs in horses is an ex vivo inflammatory model whereby blood samples collected post drug administration are stimulated with lipopolysaccharide (LPS) and calcium ionophore (CI) to induce the arachidonic acid cascade [38, 39, 40]. As the stimulated samples have drug in them, the idea is that depending on the concentration of drug at the time of the sample collection, there would be some inhibition of production of inflammatory mediators, produced by the arachidonic acid cascade. One of the advantages to this type of ex vivo model is that concentrations of the drug in the samples represent actual concentrations. In other words, the drug has been processed by the animal (undergone ADME processes including active metabolites). A second advantage is that samples are stimulated with LPS and CI, outside of the animal, avoiding potential adverse systemic effects of these two compounds.

3. Importance of Pharmacokinetic Studies

In addition to generating concentration data that can be used to establish regulatory recommendations, as mentioned earlier, PK studies can tell us a lot about the behavior and pharmacology of a drug.

3.1. Understanding Absorption

Drugs come in variety of formulations, including tablets, capsules, powders, suspensions, and solutions. The way in which a drug is formulated affects its rate of absorption, which in turn can affect the detection time of a drug, simply because the drug must reach the circulation to be eliminated from the body. Pharmacokinetic studies allow us to study how the formulation impacts the rate of absorption and the effect on detection time. Following oral administration, “solid” dosing forms must undergo dissolution and disintegration to be absorbed. These are rate‐limiting processes that can prolong the rate of absorption and potentially the detection time of a drug. Some drugs are formulated as esters to prolong their release. Procaine and benzathine are examples of compounds that are added to a drug for this purpose, essentially creating an extended‐release form of the drug, whereby the rate of absorption is slower than the rate of elimination (“flip‐flop” kinetics). The clinical benefit is that these formulations necessitate less frequent administration; however, the slower rate of absorption ultimately leads to a longer detection time.

3.2. Understanding Metabolism

Pharmacokinetic studies are also useful tools to describe the metabolism of a drug, a process essential to drug elimination. As some drugs are eliminated from the body as the administered drug, as metabolites or as a combination of both, PK studies can be useful in determining what compounds to monitor for in regulatory samples. For example, acepromazine is very effectively metabolized to 2 metabolites, called 2‐(1‐hydroxyethyl) promazine (HEP) and 2‐(1‐hydroxyethyl) promazine sulfoxide (HEPS). In the case of acepromazine and the HEPS metabolite, concentrations fall below the level of detection by 36 h post‐administration, while HEP is detectable in blood samples for 72 h [41]. With a recommended detection time of 48 h by most regulatory bodies, regulation of acepromazine is based on HEP.

4. Conclusion

The use of therapeutic substances is necessary to effectively treat equine athletes; however, these compounds must be used judiciously to ensure the welfare of the horse and integrity of horseracing. Well‐designed PK and PD studies are used to establish regulatory recommendations and can aid regulators and veterinary practitioners in this endeavor. It is also imperative, however, that the treating veterinarian do his or her own risk assessment based on relevant clinical factors.

Conflicts of Interest

The authors declare no conflicts of interest.

Knych H., “Administration Studies in Equine Antidoping Research: Designing Scientific Investigations to Effectively Direct Medication Control in Racehorses,” Drug Testing and Analysis 17, no. 9 (2025): 1560–1566, 10.1002/dta.3857.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.