Medication control of flunixin in racing horses: Possible detection times using Monte Carlo simulations

Summary Background For medication control in several jurisdictions, withdrawal time is the period of refrain from racing after drug administration. It is set by adding a safety period to an experimental detection time. However, there are no reports of statistical analyses of detection time for the determination of withdrawal time in flunixin meglumine‐treated horses. Objective To analyse the population pharmacokinetics of flunixin in horses through the generation of a dataset for detection time statistical analysis and predictions via Monte Carlo simulation. Study design Experimental study. Methods Drug plasma and urine concentrations following single intravenous administration of flunixin 1.1 mg/kg bodyweight (BW) in 10 horses and multiple administration of q 24 hours for 5 days in 10 horses were measured using liquid chromatography with tandem mass spectrometry (LC‐MS/MS). Data were modelled using a nonlinear mixed effect model followed by Monte Carlo simulation. Irrelevant plasma concentration (IPC) and irrelevant urine concentration (IUC) were calculated using the Toutain approach. Detection times were obtained considering the time after the last administration for selected quantiles of 5000 hypothetical horses under the international screening limit (ISL) proposed by the International Federation of Horseracing Authorities (plasma: 1 ng/mL, urine; 100 ng/mL). Results For a regimen of 1.1 mg/kg BW q 24 hours, the IPC and IUC values were 2.0 and 73.0 ng/mL respectively. Detection times in plasma above the ISL for 90% of simulated horses were estimated as 74 hours after a single 1.1 mg/kg dose administration, 149 and 199 hours after multiple doses over 5 days at either 24‐ or 12‐hour intervals respectively. Corresponding detection times in urine were 46, 68 and 104 hours respectively. Main limitation Only female horses were investigated. Conclusions Statistical detection times for different flunixin meglumine regimens indicated a delay of detection time in plasma after multiple administrations under ISL.


| INTRODUC TI ON
Flunixin meglumine is a one of the most commonly employed NSAIDs in cases of inflammation and pain associated with soft tissue conditions in horses. [1][2][3][4] It is also considered efficacious in controlling abdominal pain, and thus is the standard therapy for equine colic. [5][6][7] Most horse racing regulatory authorities have distinct frameworks for doping control and medication control. In the case of doping drugs, such as anabolic steroids, the objective is to detect any trace through the most sensitive analytical methods. 8 concentrations are below the screening limit (SL). 11 Detection time experiments are usually conducted after the administration of a single dose in six to eight horses. However, detection time is only preliminary information provided without any statistical basis. As described by the EHSLC or the Fédération Equestre Internationale, detection time depends on various factors, including the dosage regimen, route of administration, pharmaceutical formulation, breed, age, sex and, most importantly, the number of investigated horses. The last factor has been explored through a series of Monte Carlo simulation studies in order to allow prescribers to set a withdrawal time for a particular horse, including an additional safety period which will be even longer than detection times observed for a limited number of horses. 11 Although the pharmacokinetics of flunixin in horses have already been established, 12

| Study design
Twenty healthy 3-to 10-year-old experimental Thoroughbred female horses with a bodyweight (BW) of 428-530 kg were used.
Horses were examined by a veterinarian and found healthy prior to the investigation. Horses were kept in individual stalls during experiments and had ad libitum access to grass, hay and water. Straw was completely replaced every day.
Flunixin meglumine dose (1.1 mg/kg BW flunixin) was determined based on a previous report. 13 Flunixin meglumine was administered into the right jugular vein via a short bolus infusion (<10 seconds).
Plasma and urine samples were stored at −20°C until analysis.

| Sample analysis
Twenty microlitres of methanol (containing 1 μg/mL diclofenac-d4 [Toronto Research Chemicals] as an internal standard) and 0.1 mL 1 mol/L sodium hydroxide were added to 0.1 mL plasma or urine, and the mixture was incubated at room temperature for 10 min. To hydrolyse urine, 1 mL 1 mol/L acetate buffer (pH 5.0) and 4 mL tert-butyl methyl ether were added, and the mixture was stirred for 5 minutes.
The upper organic phase was dried under a nitrogen stream at 40°C.
The residue was reconstituted with 0.5 mL of 0.1%(v/v) formic acid

| Pharmacokinetic analysis
Plasma pharmacokinetic analyses were conducted using a nonlinear mixed effect (NLME) model on commercially available software To report interindividual variability as a coefficient of variation, Equation (2) was used for the conversion of variance terms (ω 2 ) into a coefficient of variation (CV%).
Shrinkage of the random effects (eta) towards the means was  The additive sigma was reported as its standard deviation noted with the same units as plasma concentration (µg/mL), and the multiplicative sigma was reported as a coefficient of variation. The precision of the parameters was estimated using a bootstrap tool (n = 50 replicates).
Plasma clearance was used to calculate the effective plasma concentration (EPC) via Equation (5). 10 (1) parameter_i = tv_parameter ⋅ exp( i ), In the US, the Racing Medication and Testing Consortium (RMTC) compute withdrawal time with its confidence interval as done for drug residues in food-producing animals. 17 The RMTC SL for plasma flunixin is 5 ng/mL. We also generated a meta-population of 5000 detection times for this US SL in addition to computing the 95% confidence interval of the 95th percentile considered by the RMTC via the nonparametric bootstrap tool of Crystal Ball (Oracle Corporation), which is appropriate for estimating the reliability of forecast statistics.

| RE SULTS
Semilogarithmic plots of the disposition curves for plasma and urine flunixin concentrations in each horse are depicted in Figure 1.
Descriptive statistics characterising flunixin elimination in plasma and urine are given in Table 1 Check ensured that simulated data were consistent with observed data (Figure 4).
The interindividual variability computed with Equation (2) and post-hoc values are given in Table 2. The variability for Rss was highest, with a coefficient of variation of 55.5% corresponding to Rss post-hoc values from 6.0 to 97.1. Bootstrap estimates of the primary structural parameters (thetas), secondary parameters and their associated coefficients of variation, as a measure of the precision of their estimation, are given in

| D ISCUSS I ON
The pharmacokinetics of flunixin were previously reported in several studies, 13,14,18,19 but its population pharmacokinetics in horses remain to be established. The average steady-state volume of distribution and clearance reported were 0.137-0.157 L/kg and 0.046-0.062 L/kg/h, respectively, similar to our results. 13,14 The numerical value of the Rss is rarely mentioned in the literature, but the average Rss value obtained herein (Rss = 37) seems consistent with those deduced from published raw data or corresponding flunixin depletion curves. 13 For convenience of urine collection, only female horses were used in this study. We believe that this has no impact on the generality of our results as no sex differences in flunixin disposition have been reported. Flunixin clearance was significantly lower in old compared with young horses. 12 To prevent possible age-related bias, we selected equids representative of the racehorse population. is assumed that the EPC should be close to the EC50. In the PK/PD analysis of the clinical effect of stride length and rest angle in experimental arthritis, the EC50 of flunixin was reported to range from 0.24 to 0.93 µg/mL, which was close to the EPC computed herein (0.96 µg/mL). 23 The IPC and IUC calculated in this study were similar to the ISL of IFHA.
Several studies have explored flunixin's effects on thromboxane B2 production in horses, as thromboxane generation is considered an index of NSAID efficacy. 13,23-25 A single 1.1 mg IV flunixin administration maintained a significant effect for 24 hours, which was lost when the plasma concentration decreased to 10-20 ng/mL. 13,25 Considering a PK/PD factor of 50, the average IPC can be estimated to 18.9 ng/mL, which is close to the concentration at which the significant effect on thromboxane production was compromised, indicating the validity of EPC and IPC.
Detection time experiments are often conducted after a single dose. It should be considered that detection time depends on various biological and clinical factors. These include between-subject variability of PK parameters, and differences in administration route, dosage or formulation. It has already been reported that multiple doses may prolong detection time. 26,27 In this study, 10 horses were administered flunixin q 24 hours for 5 days, and data were subjected to Monte Carlo simulation. Veterinarians can select the withdrawal time based on the results provided in Table 5.  30 Such an amount can be partly recycled when the horse ingests litter straw contaminated with its own urine as reported for flunixin and meclofenamic acid. [31][32][33] Thus, we replaced the straw in stalls daily, preventing a prolonged detection time due to flunixin recycling.
When considering the IFAH ISL, detection times in plasma were longer than in urine. This is related to the Rss difference used by the IFHA when determining the plasma and urine ISL (ie 100) and the one computed in this study (ie 37). Of note, the urine-to-plasma ratio is subject to considerable variation. In our horses, the interindividual variability of Rss was 55.5%, with values ranging from 6 to 97. The urine-to-plasma ratio depends on various factors, such as diet and water intake, rendering urine less attractive for robust medication control. 10 The alternative to the EHSLC approach followed herein is that of RMTC in the USA, which consists of calculating statistical withdrawal times as for drug residues in food-producing animals.
This method guarantees that a positive result to be highly unlikely. In the formal context of a Risk Analysis for medication control as described by Toutain, 9 the first step is the Risk Assessment, identical between EU and US approaches and requiring the generation of robust data to be analysed via the advanced scientific tools, such as population modelling. 9 It is only during the second step, Risk Management, that the same data and results can be handled differently, leading to different recommendations based on the values specific to each jurisdiction. Within this framework of Risk Assessment, we generated data on flunixin meglumine which could be utilised in the approaches described above.
In conclusion, the flunixin IPC and IUC calculated for 20 horses were consistent with the IFHA ISL. MCS indicated that a detection time of 144 hours, as proposed by EHSLC and IFAH, is appropriate for a single flunixin meglumine administration at the assessed dose.
However, the delay after multiple administrations may not be sufficient to ensure negativity, especially in plasma. This study provides statistical detection times which should facilitate the determination of individual withdrawal times by clinicians.

E TH I C A L A N I M A L R E S E A RCH
All experiments were approved by the Animal Care and Use Committee of Equine Research Institute, Japan Racing Association and Laboratory of Racing Chemistry.

CO N FLI C T O F I NTE R E S T S
No competing interests have been declared. Toutain contributed to pharmacokinetic analysis and manuscript preparation.

I N FO R M E D CO N S E NT
Not applicable.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1111/evj.13532.