Skip to main content
NCBI home page
Bioanalysis logoLink to Bioanalysis
. 2024 Sep 5;16(17-18):947–958. doi: 10.1080/17576180.2024.2385848

Quantification of osilodrostat in horse urine using LC/ESI–HRMS to establish an elimination profile for doping control

Hideaki Ishii a,b,*, Ryo Shigematsu a, Shunsuke Takemoto a, Yuhiro Ishikawa c, Fumiaki Mizobe c, Motoi Nomura c, Daisuke Arima d, Hirokazu Kunii d, Reiko Yuasa d, Takashi Yamanaka e, Sohei Tanabe e, Shun-ichi Nagata a, Masayuki Yamada a, Gary Ngai-Wa Leung a
PMCID: PMC11486175  PMID: 39235065

Abstract

Aim: The use of osilodrostat, developed as a medication for Cushing's disease but categorized as an anabolic agent, is banned in horses by both the International Federation of Horseracing Authorities and the Fédération Equestre Internationale. For doping control purposes, elimination profiles of hydrolyzed osilodrostat in horse urine were established and the detectability of free forms of osilodrostat and its major metabolite, mono-hydroxylated osilodrostat (M1c), was investigated.

Materials & methods: Post-administration urine samples obtained from a gelding and three mares were analyzed to establish the elimination profiles of osilodrostat using a validated method involving efficient enzymatic hydrolysis followed by LC/ESI–HRMS analysis.

Results: Applying the validated quantification method with an LLOQ of 0.05 ng/ml, hydrolyzed osilodrostat could be quantified in post-administration urine samples from 48 to 72 h post-administration; by contrast, both hydrolyzed osilodrostat and M1c were detected up to 2 weeks. In addition, confirmatory analysis identified the presence of hydrolyzed osilodrostat for up to 72 h post-administration.

Conclusion: For doping control purposes, we recommend monitoring both hydrolyzed M1c and osilodrostat because of the greater detectability of M1c and the availability of a reference material of osilodrostat, which is essential for confirmatory analysis.

Keywords: : anabolic agent, doping control, elimination profile, equine, LC/ESI–HRMS, osilodrostat, quantitative analysis, urine

Plain language summary

Article highlights.

Background

  • The use of osilodrostat, developed as a medication for Cushing's disease but categorized as an anabolic agent, is banned in horses by both the International Federation of Horseracing Authorities and the Fédération Equestre Internationale.

  • For doping control purposes, we conducted administration studies of osilodrostat using one gelding and three mares.

  • We previously reported the excretion profiles of osilodrostat and its major metabolite, mono-hydroxylated osilodrostat (M1c), in horse plasma as well as the establishment of the metabolic pathway of osilodrostat in equine urine. The profiles indicated that osilodrostat was predominately excreted into urine as M1c and osilodrostat N-glucuronide (M2), together with several phase II conjugates of osilodrostat and M1c.

  • In this study, elimination profiles of hydrolyzed osilodrostat in horse urine were established and the detectability of free forms of osilodrostat and M1c was investigated.

Methodology for enhancing the detectability of targeted analytes

  • The hydrolysis procedure for the phase II conjugates of osilodrostat and M1c in horse urine was optimized to enhance their detectability.

  • Post-administration urine samples obtained from a gelding and three mares were analyzed to establish the elimination profiles of osilodrostat using a validated method involving efficient enzymatic hydrolysis followed by LC/ESI–HRMS analysis.

  • The optimized enzymatic hydrolysis method showed high efficiency, with more than 90% of the four different forms conjugates of each of osilodrostat and M1c.

  • The quantification method of hydrolyzed osilodrostat in urine in the range 0.05–15 ng/ml was successfully validated.

Elimination profiles of osilodrostat in horses & investigation of detectability of osilodrostat & its major metabolite based on the qualitative analyses

  • The maximum urinary concentrations of osilodrostat and M1c reached at 3 h post-administration for all four horses and their concentrations were gradually decreased following first-order elimination kinetics.

  • Hydrolyzed osilodrostat could be quantified from 48 h to 72 h post-administration for four horses, with an LLOQ of 0.05 ng/ml; by contrast, both hydrolyzed osilodrostat and M1c were detected up to the last sample collection time of 2 weeks.

  • For confirmation of osilodrostat using three diagnostic product ions, its presence could be identified for up to 72 h and 48 h post-administration for the three mares and the gelding, respectively.

Recommended monitoring substances for doping control

  • For doping control purposes, we recommend monitoring both hydrolyzed M1c and osilodrostat because of the greater detectability of M1c and the availability of a reference material of osilodrostat, which is essential for confirmatory analysis.

1. Introduction

Osilodrostat (also known as LCI699) is an inhibitor of 11β-hydroxylase (CYP11B), which catalyzes the biosynthesis of adrenal cortisol in the adrenal cortex and is used for the treatment of patients with Cushing's disease, which is caused by the overproduction of cortisol [1,2]. Osilodrostat was approved by the U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) and was launched onto the market in 2020 as an orally available prescription medication with the brand name Isturisa [2]. However, osilodrostat has been exemplified in the prohibited list under the category of “other anabolic agents” by the World Anti-Doping Agency (WADA) since 2022 [3] because osilodrostat was reported to upregulate the biosynthesis of endogenous testosterone resulting from compensatory stimulation of ACTH triggered by the decreased levels of endogenous cortisol [1]. This phenomenon is believed to be particularly remarkable in females because approximately one-half of the testosterone in females is generated in the adrenal gland, whereas in males, testosterone is predominately biosynthesized in the Leydig cells (testosterone-producing cells) in the testes [1]. In light of this anabolic effect, the administration of osilodrostat to horses in both horseracing and equestrian sports brings a lifetime ban by the International Federation of Horseracing Authorities (IFHA) [4] and the Fédération Equestre Internationale (FEI) [5], respectively. Given the ever-increasing availability of osilodrostat as a pharmaceutical product and reagent [1,2], there is a risk of its abuse and/or misuse in equines [3].

The pharmacokinetics of osilodrostat in humans and preclinical species (e.g., rodent) has been well investigated and elucidated through clinical and preclinical studies [1,2,6–9]. In humans [1,2], the major metabolites of osilodrostat in plasma were determined to be di-oxygenated osilodrostat (51% of total radioactivity) and osilodrostat N-glucuronide (9%) after a single oral administration of 14C-labeled osilodrostat at a dose of 50 mg (3.7 MBq). In urine, the major metabolites were found to be osilodrostat N-glucuronide (17%), di-oxygenated osilodrostat glucuronide (13%) and mono-hydroxylated osilodrostat (11%). However, the literature contains no metabolism study of osilodrostat in horses. Therefore, we conducted a pharmacokinetic study of osilodrostat using one gelding and three mares, each of which received an oral dose of 50 mg via a nasoesophageal tube [10,11]. We have also reported a generic approach for the identification of drug metabolites on the basis of differential analysis between pre- and post-administration samples using high-resolution mass spectrometry (HRMS) data [12–15]. This approach was applied to the pharmacokinetic study of osilodrostat in horses to screen potential metabolites in plasma and urine [10,11]. Our results showed that mono-hydroxylated osilodrostat (M1c) was the predominant metabolite in plasma, followed by the parent drug of osilodrostat and that the elimination profile of osilodrostat in plasma was successfully established [10]. For urine analysis using differential analysis, we established a metabolic pathway of osilodrostat in horse urine for the first time, identifying 37 metabolites [11]. In addition, we found that osilodrostat was predominantly metabolized into M1c (∼40%) and osilodrostat N-glucuronide (M2) (∼17%), together with several conjugates of either osilodrostat or M1c (Figure 1) [11]. Given the presence of these phase II metabolites, we recommend detecting osilodrostat and M1c after applying an appropriate hydrolysis method to extend their detection periods as much as possible.

Figure 1.

Figure 1.

Open in a new tab

Brief summary of the metabolic pathway of osilodrostat in equines. Percentages represent approximate relative compositions of osilodrostat and its metabolites in the post-3 h administration samples after 50 mg of osilodrostat was administered to four horses. An entire osilodrostat metabolic pathway including 37 metabolites is available in the authors' previous study [11].

In the present study, we developed and validated a quantification method for osilodrostat in horse urine, employing an efficient enzymatic hydrolysis of both osilodrostat conjugates and M1c conjugates using liquid chromatography electrospray ionization HRMS (LC/ESI–HRMS). To establish elimination profiles of osilodrostat in horse urine, we successfully applied the method to post-administration urine samples obtained from a gelding and three mares after oral administration of osilodrostat (50 mg) via a nasoesophageal tube. The detectability of osilodrostat and M1c in these specimens were investigated on the basis of the corresponding qualitative analyses.

2. Materials & methods

2.1. Materials

Osilodrostat and its internal standard (IS) of 4-(5,6,7,8-tetrahydroimidazo [1,5-a] pyridin-5-yl) benzonitrile were purchased from Cayman (Ann Arbor, MI, USA) and Ambeed (Arlington Heights, IL, USA), respectively. Acetic acid, sodium acetate, LC/MS-grade formic acid, LC/MS-grade methanol, LC/MS-grade acetonitrile and 28 w/v% ammonia solution were obtained from Fujifilm Wako Pure Chemical (Osaka, Japan). Ammonium formate was acquired from Sigma-Aldrich Chemical (St. Louis, MO, USA). Four different types of β-glucuronidase were used. β-Glucuronidase from Pomacea canaliculata (22,000 Fishman U/ml) and Escherichia coli K 12 were acquired from Nippon Bio-Test Laboratories (Saitama, Japan) and Roche Diagnostics (Mannheim, Germany), respectively, whereas β-glucuronidase from genetically selected Haliotis rufescens (β-glucuronidase activity: >100,000 U/ml, sulfatase activity: <8,000 U/ml) and BG turbo (recombinant β-glucuronidase, its activity: >200,000 U/ml) were obtained from KURA Biotech (Los Lagos, Chile). Hydrogen chloride in methanol (1 mol/l) was purchased from Kokusan Chemical (Tokyo, Japan). Solid-phase extraction (SPE) strong cation-exchange (SCX) cartridge columns (Oasis MCX, 3 ml, 60 μm) and centrifuge tubes with polytetrafluoroethylene (PTFE) filters (0.45 μm) were purchased from Waters (Milford, MA, USA) and Millipore (Bedford, MA, USA), respectively. High-purity water was generated with a Milli-Q purification system (Millipore).

2.2. Administration study

All experimental procedures were approved by the Animal Welfare and Ethics Committee of the Equine Research Institute of the Japan Racing Association (authorization number 23–31) for a pilot study and by the Laboratory of Racing Chemistry (authorization number 23–04) for the main study as reported in the authors' previous report [10,11]. A gelding (4 years old, 501 kg) kept in the Equine Research Institute (Shimotsuke, Japan) and three mares (5.3 ± 2.5 years old, 478 ± 33 kg) bred at the Horse Racing School (Shiroi, Japan) were used for the pilot study and the main study, respectively. Ten tablets of Isturisa 5 mg (Recordati Rare Diseases Japan, Tokyo, Japan) were homogenized with a mortar and pestle and the resultant powder was suspended in 500 ml of water; the suspension was orally administered to each horse via a nasoesophageal tube followed by flushing with an additional 2 L of tap water. Urine samples from the gelding were collected at 0 (pre), 3, 6, 9, 24, 48, 120, 192 and 336 h to assess potential toxicity as part of the pilot study, whereas samples from the three mares were collected at 0 (pre), 3, 6, 9, 24, 30, 48, 54, 72, 120, 168, 240 and 336 h. After sample collection, urine samples were stored in a deep freezer at -80°C prior to analysis.

2.3. Preparation of stock & working solutions

Standard and IS stock solutions at a concentration of 1.0 mg/ml were individually prepared in methanol. Standard working solutions of osilodrostat at concentrations of 0.05, 0.15, 0.5, 1.5, 5 and 15 ng/ml were prepared by serially diluting the stock solution with 1 mol/l sodium acetate buffer (AB) at pH 5.0. The IS working solution at 5 ng/ml was also prepared by diluting the stock solution with AB.

2.4. Preparation of quality control samples

To cover the entire calibration range, four representative levels of quality control (QC) samples, including the lower limit of quantification (LLOQ), low QC (LQC), medium QC (MQC) and high QC (HQC), were prepared as follows. One hundred microliters of standard working solutions at 5, 15, 100 and 1200 ng/ml were diluted 100-fold with pooled blank urine to achieve final concentrations of 0.05, 0.15, 1 and 12 ng/ml respectively.

2.5. Sample preparation

A blank sample was prepared by mixing 100 μl of blank pooled urine and 200 μl of AB. For calibration curve samples in the concentration range 0.05–15 ng/ml, 100 μl of each standard working solution and 100 μl of the IS working solution were added to each 100 μl of blank pooled urine. To prepare QC and test samples, 100 μl of AB and 100 μl of the IS working solution were spiked into 100 μl of urine to ensure a consistent composition of the matrix and solvent across all samples including blanks, calibrators, QC samples and test samples. All samples were then spiked with 400 μl of β-glucuronidase and 1.3 ml of AB (total volume of 2 ml) and incubated at 60°C at 150 rpm for 24 h, followed by centrifugation (2000 × g, 4°C, 5 min). After the SPE column (Oasis MCX; 3 ml) was preconditioned with methanol (3 ml) and water (3 ml), the obtained supernatant was passed through the column. The column was washed with 3 ml of 2% formic acid and 3 ml of methanol, followed by the elution of the analytes and the IS with 28% ammonia/methanol (5: 95) into a clean tube. The eluate was evaporated to dryness under a stream of nitrogen gas. After reconstitution of the residue in 100 μl of 10 mmol/l ammonium formate, the reconstituted solution was filtered with a centrifuge tube with a PTFE filter (0.45 μm) at 10,000 × g at 4°C for 3 min. Finally, 10 μl of the filtrate was injected into an LC/ESI–HRMS system. If required, the test samples with urinary concentrations of osilodrostat expected to be greater than the upper limit of quantification were diluted using a blank matrix with dilution factors of 10, 50 and 100, accordingly, prior to extraction.

2.6. Instrumental conditions

Analyses were carried out using a Vanquish UHPLC system series liquid chromatograph linked to a quadrupole Orbitrap high-resolution mass spectrometer (Orbitrap Exploris 240) equipped with a heated ion-spray interface (Thermo Fisher Scientific, Waltham, MA, USA).

The analytical conditions for quantitative analysis were as follows. Analytes and IS were separated on an ACQUITY UPLC BEH C18 column (2.1 mm i.d. × 100 mm, 1.7 μm, Waters) at 40°C. The binary mobile phase consisting of solvent A (10 mmol/l ammonium formate in water) and solvent B (acetonitrile) was employed at a flow rate of 0.6 ml/min, with the following gradient elution: 0–3.5 min, 10–50% B; 3.51–4.5 min, 98% B; and 4.51–5.5 min, 10% B.

Optimized HRMS setting parameters were as follows: sweep gas, 10 arbitrary units; auxiliary gas, 20 arbitrary units; sheath gas, 40 arbitrary units; spray voltage, 3.5 kV for positive ESI; in-source collision-induced dissociation (CID), 0 V; vaporizer temperature, 400°C; S-lens voltage, 80.0%; ion transfer tube temperature, 400°C; collision gas, nitrogen; and scan range, m/z 50–750. For full MS, maximum injection (IT) time, auto; auto gain control (AGC) target, standard; and orbitrap resolution, 60,000. For product ion scan, maximum IT, auto; AGC target, standard; orbitrap resolution, 30,000; Q1 resolution, 0.7 m/z; run time, 0.5–4.5 min; precursor ions, m/z 228.0932 and m/z 224.1182 for osilodrostat and IS, respectively; product ions, m/z 81.0447 and m/z 82.0525; CE, 30 eV for both the analyte and IS. All peak integrations and related calculations for qualitative and quantitative analyses were performed using the Trace Finder software (version 4.1, Thermo Fisher Scientific). The plot of each peak area ratio (analyte/IS) as a function of the corresponding nominal concentration of osilodrostat was fitted by least-squares linear regression using 1/y2 weighting factors.

Qualitative analysis was performed using the same analytical conditions as the quantitative analysis described above, with minor modifications to the following gradient conditions and product ion scan conditions. The gradient elution was conducted as follows: 0–2 min, 2% B; 2–16 min, 2–90% B; 16–20 min, 90–98% B; 20–22 min, 98% B; and 22.01–24 min, 2% B. The product ion scan conditions for osilodrostat were as follows: precursor ion, m/z 228.0932; product ions, m/z 81.0447, m/z 54.0338 and m/z 134.0401; collision energy, 30 eV. Those for M1c were as follows: precursor ion, m/z 244.0881; product ions, m/z 97.0396, m/z 95.0604 and m/z 226.0775; collision energy, 20 eV.

2.7. Method validation

As method validation parameters, the linearity, selectivity, intra- and inter-assay reproducibility, matrix effect and recovery were evaluated. For linearity, the accuracy (%) of the back-calculated concentration with respect to the corresponding nominal concentration was determined using five datasets on different days. For selectivity, a total of six different urine samples obtained from three male and three female horses were analyzed. For reproducibility, the four levels of QC samples prepared in the section “Sample preparation” were analyzed in six replicates on three different days to determine the intra-assay and inter-assay accuracy and precision. The criteria for accuracy and precision (expressed as the coefficient of variance [CV%]) were within ±15% (±20% at the LLOQ) and less than or equal to 15% (≤20% at the LLOQ). For the matrix effect, we investigated matrix factors by comparing the MS signals at the two spiked concentrations (i.e., the LQC and HQC) in the presence and absence of the matrix using the aforementioned six different individual urine samples. Extraction recovery from urine was determined at three QC levels (i.e., LQC, MQC and HQC) and was calculated by comparison of the peak area ratios (analyte/IS) between pre-spiked (extracted) samples and post-spiked (unextracted) samples. In all cases, IS was spiked into the final extracts. The recovery of IS at 1 ng/ml was determined in a similar manner.

2.8. Detection periods of potential monitoring substances in urine

For the purpose of doping control, the detection periods of analytes based on the qualitative analysis are critical for interpreting a positive case. Thus, the extracts prepared as described in the “Sample preparation” section for quantitative analysis were re-analyzed using the method for qualitative analysis described in the “Instrumental condition” section to determine the detection periods of osilodrostat and M1c. In addition, the limit of detection (LOD) and limit of confirmation (LOC) for osilodrostat were experimentally determined by analyzing the spiked samples at 0.5, 1, 2, 5, 10, 20, 50, 150 and 500 pg/ml in triplicate. The LOD is defined as the minimum concentration at which an analyte of interest can be detected with a signal-to-noise (S/N) ratio greater than three in triplicate, whereas the LOC is the lowest level at which relative abundances of three diagnostic product ions of an analyte in a standard solution are identical to those in a test sample, as stipulated in the “AORC Guidelines for the Minimum Criteria for Identification by Chromatography and Mass Spectrometry” published by the Association of Official Racing Chemists (AORC) [16].

3. Results

3.1. Method development

A quantification method for osilodrostat in urine was developed with modifications of our validated method for plasma [10] after mainly incorporating an enzymatic hydrolysis procedure because we previously reported that osilodrostat mainly existed as the N-glucuronide form (M2), together with three other types of conjugates of glucose (M3), ribose (M4) and riburonic acid (M5) [11]. This procedure would also be effective to firmly extend the detection times of mono-hydroxylated osilodrostat in horse urine because four of its different conjugates (i.e., two glucuronides and two glucosides) have been reported although their relative amounts were small (a total of ∼2%) [11]. Thus, four hydrolysis procedures—alkaline, acidic and enzymatic hydrolysis as well as solvolysis—were attempted and the hydrolysis efficiency was calculated by comparing the peak areas of each phase II conjugate of osilodrostat and M1c in the post-3 h administration urine sample with and without hydrolysis. When alkaline hydrolysis with 1 mol/l sodium hydroxide at 60°C for 1 and 2 h and methanolysis with 1 mol/l hydrochloride in methanol at 60°C for 40 min were applied, degradation of more than 99% of the osilodrostat was observed. By contrast, incubation of the post-administration sample with hydrochloride solution at 1, 2, 3 and 6 mol/l at 60°C for 1 h resulted in poor hydrolysis efficiencies of approximately 20–30% for both conjugates of osilodrostat and M1c.

Then, to 100 μl of the post administration urine, the use of 50 μl of β-glucuronidase solutions from four different sources of P. canaliculata, recombinant, H. rufescens and E. coli K 12 were screened at 60°C for 3 h, giving 7.2-, 5.6-, 4.6- and 1.3-fold increases in the MS response of osilodrostat, respectively. Thus, the glucuronidase solution from P. canaliculate was used. Eventually, more than 92% and 95% hydrolysis efficiencies were confirmed for four conjugates of osilodrostat (glucuronide, glucoside, ribose conjugate and riburonic acid conjugate) and four conjugates of M1c (two glucuronides and two glucosides), respectively, under the conditions described in the section “Sample preparation.” The final increases in the peak areas of osilodrostat and M1c were 31.6-fold and 1.25-fold, respectively.

For the remaining procedures, including a sample extraction followed by instrumental analysis, approximately the same procedures were applied as those applied in the quantification method for plasma [10] after some minor modifications were made to, for example, the composition of the reconstituted solution.

3.2. Method validation

Validation results of the quantification method for osilodrostat in urine equipped with hydrolysis are summarized in Table 1. For the linearity of five calibration curves, all of the accuracies in the concentration range 0.05 ng/ml to 15 ng/ml were between 96.0 and 111%, with coefficients of determination greater than 0.9971. For selectivity, no interfering peaks for osilodrostat and its IS were observed at their expected retention times in six different lots of horse urine. For reproducibility, the accuracies of intra- and inter-assay at four QC levels in six replicates on three different days were 88.4–115% and 90.1–98.7%, respectively, whereas the precisions were less than or equal to 5.9% and 12.2%, respectively. For the matrix effect, relatively high ionization suppressions for osilodrostat and the IS with matrix factors of 0.204–0.253 and 0.307–0.352, respectively and their CV values were ≤4.4% among LQC and HQC samples prepared from the six individual urine samples.

Table 1.

Summary of validation results of the quantification method for osilodrostat in equine urine.

Validation parameters (Condition) Spiked concentration (ng/ml)   Osilodrostat IS
Calibration curve (n = 5) 0.05 Accuracy (%) 100–101 N/A
  0.15–15   96.0–111 N/A
  0.05–15 R 2 ≥0.9971 N/A
Selectivity (6 individuals) 0.05, 1 (IS) Area ratio (%) 0 0
    (LLOQ vs blank)    
Intra-assay reproducibility 0.05 Accuracy (%) 90.3–115 N/A
(4 conc., n = 6)   CV (%) 2.3–4.8 N/A
  0.15, 1, 12 Accuracy (%) 88.4–103 N/A
    CV (%) 1.4–5.9 N/A
Inter-assay reproducibility 0.05 Accuracy (%) 98.7 N/A
(4 conc., n = 6, 3 days)   CV (%) 12.2 N/A
  0.15, 1, 12 Accuracy (%) 90.1–98.5 N/A
    CV (%) 2.4–5.9 N/A
Matrix effect (2 conc., 6 individuals) 0.15, 12, 1 (IS) Matrix factor 0.204–0.253 0.307–0.352
    CV (%) 3.1–3.6 1.7–4.4
Recovery (3 conc., n = 3) 0.15, 1, 12, 1 (IS) Recovery (%) 84.5–85.8 82.3
Open in a new tab

N/A: Not applicable.

3.3. Application of the quantification method to the administration study

After the method validation for osilodrostat in horse urine was completed, the quantitative method was successfully applied to the post-administration urine samples. Representative data of extracted ion chromatograms of osilodrostat and its IS in horse urine are depicted in Figure 2. Osilodrostat and the IS were detected at retention times of 2.8 min and 3.1 min, respectively, from a calibrator at 0.05 ng/ml and the post-72 h administration sample (Figures 2A & C), whereas the absence of osilodrostat in the pre-administration urine sample was confirmed (Figure 2B).

Figure 2.

Figure 2.

Open in a new tab

Typical extracted ion chromatograms of osilodrostat and its IS in horse urine. Left: Osilodrostat; right, IS (TIPB); (A) Spiked sample at 0.05 ng/ml of osilodrostat in urine; (B) Pre-administration urine sample; (C) Post-administration urine sample at 72 h. product ion scan conditions for osilodrostat and IS, respectively: precursor ions, m/z 228.0932, m/z 224.1182; collision energies, 20 eV, 30 eV; product ions, m/z 81.0447, m/z 82.0525.

The elimination profiles of osilodrostat in horse urine were successfully established (Figure 3). The urinary maximum concentrations reached 444 ng/ml for a gelding and 761 ± 56 ng/ml (mean ± standard deviation) for three mares, respectively, at post-3 h administration. Both concentrations gradually decreased over time with similar first-order elimination kinetics. As a result, osilodrostat could be quantified up to post-48 h administration for the gelding, post-54 h for one mare and post-72 h for two mares. Notably, the post-54 h and -72 h administration samples for the gelding were unavailable because of the fixed protocol used in the pilot study.

Figure 3.

Figure 3.

Open in a new tab

Elimination profile of osilodrostat in urine after 50 mg of osilodrostat was orally administered to a gelding and three mares via a nasoesophageal tube.

Note that post-54 h and -72 h administration samples for the gelding were unavailable according to the predefined protocol for a pilot study.

3.4. Detectability of osilodrostat & M1c in post-administration samples

The samples prepared as described in the “Sample preparation” section were re-analyzed using a different LC/ESI–HRMS method, as described in the section “Instrumental conditions.” The base peaks of both osilodrostat at m/z 81.0447 and M1c at m/z 97.0396 were detected with an S/N greater than 3 in all post-administration samples collected from 3 h to 336 h (i.e., 2 weeks) after administration. In addition, the LOD and LOC of osilodrostat in horse urine were experimentally estimated to be 10 pg/ml and 150 pg/ml, respectively.

For confirmatory analysis in horse racing, the consistency of retention times and the relative abundances of at least three diagnostic product ions between the test sample(s) and the reference material is required as stipulated in the AORC guideline for the minimum criteria for identification by chromatography and mass spectrometry [16]. By applying the aforementioned criteria, we identified osilodrostat in the post-administration samples for as long as 72 h for the three mares (Figure 4B) and 48 h for the gelding (note that the post-54 h and -72 h administration samples for the gelding were unavailable). However, beyond these periods, one or two product ions at m/z 54.0338 and m/z 134.0401 could not be detected because of insufficient amounts of hydrolyzed osilodrostat in the post-administration urine samples.

Figure 4.

Figure 4.

Open in a new tab

Overlay of extracted ion chromatograms of osilodrostat and mono-hydroxylated osilodrostat with their three diagnostic product ions. Left, osilodrostat; right, mono-hydroxylated osilodrostat (M1c). (A) 50 pg of osilodrostat reference material; (B) 72-h post-administration urine sample; (C) 3-h post-administration urine sample with 100-fold dilution; (D) 240-h post-administration urine sample. Polarity, positive; mass accuracy, 5 ppm; precursor ion of osilodrostat, m/z 228.0932, precursor ion of M1c, m/z 244.0881; product ions of osilodrostat, m/z 81.0447, m/z 54.0338 and m/z 134.0401, product ions of M1c, m/z 97.0396, m/z 95.0604 and m/z 226.0775. RA, relative abundance (%) (ratio of the peak area of the product ion to the peak area of the base peak).

In a positive case of osilodrostat in equine urine, evidence of the detection of its urinary metabolite(s) would strongly support the conclusion that the horse was exposed to osilodrostat. In light of this, despite the lack of a reference material, we investigated the detectability of three diagnostic product ions of the major urinary metabolite M1c and the use of three diagnostic product ions aimed to increase the confidence in the identification of M1c. The results show that the three product ions were clearly detected in the post-3 h administration sample (Figure 4C) and that reproducible detections of M1c in terms of relative abundances of the three product ions were barely observed up to 192 h for the gelding and 120, 168 and 240 h for the three mares, respectively (Figure 4D). After such periods, one or two characteristic product ions of M1c at m/z 95.0605 and m/z 226.0777 disappeared as the excreted amount of M1c in the urine decreased over time.

4. Discussion

To establish the elimination profiles of osilodrostat in horse urine and effectively extend the detection periods of osilodrostat and its major metabolite (M1c), enzymatic hydrolysis of their conjugates with β-glucuronidase was employed because of superior hydrolysis efficiency over other three hydrolysis procedures with different principals. However, the relatively large amount of β-glucuronidase solution was required for the completion of hydrolysis of the conjugates, resulting in relatively high ionization suppressions of the analyte and its IS. Nevertheless, given the degree of ionization suppression of the analyte and IS was relatively constant, the effect of the matrix on the quantification values was deemed negligible.

After applying the validated quantification method to the post-administration urine samples, almost same elimination curves of hydrolyzed osilodrostat in one gelding and three mares over time with similar first-order elimination kinetics were observed, suggesting no substantial gender difference in their elimination rates (slopes of urinary concentration–time curves).

In addition to quantification analysis, qualitative analyses based on screening and confirmatory tests also provided beneficial information about the detection windows in terms of doping control of osilodrostat in horses. However, as previously reported [11], osilodrostat was metabolized into four different phase II conjugates of osilodrostat (M2, M3, M4 and M5) and three types of mono-hydroxylated osilodrostat (M1a, M1b and M1c). When these phase II conjugates of a parent drug are analyzed by LC/ESI–HRMS, a conjugate peak is observed along with the parent peak at the same retention time as the conjugate peak [12–15]. This phenomenon arises because the parent drug is generated in the ion source after deconjugation (cleavage of the conjugate to form the parent drug) caused by in-source CID during the process of ions being incorporated into a mass spectrometer [17–19]. In addition, the three metabolites of mono-hydroxylated osilodrostat sharing the same molecular mass have been observed to elute at similar retention times of 11.4, 11.6 and 12.3 min for M1a, M1b and M1c, respectively, under the LC condition with a longer gradient elution [11]. Therefore, baseline separations of target analytes and their metabolites were required for qualitative analyses (in particular for confirmatory analyses, vide supra) to obtain reproducible mass spectral results by avoiding potential interferences from such metabolites. In light of the required optimum separations among the target analytes, the different LC condition from the quantification method was employed for confirmatory analyses.

The investigation results of detectability of osilodrostat and M1c indicated that relatively shorter detection periods of three diagnostic ions in the range of 48–72 h and 120–240 h for osilodrostat and M1c, respectively, was observed compared with longer detection periods of 336 h based on the detection of their base peaks as screening analyses. Please note that a urine sample volume of only 100 μl was used, with a main objective of establishing the elimination profile of osilodrostat from the horse body. The same extract was also used to study the detection period of osilodrostat and M1c. However, if a larger volume of urine is used for analysis, we speculate that the detection periods of osilodrostat in horse urine can be easily extended.

In general, hair analysis is a common approach to extend the detectability of administered drugs. Over the past two decades, hair analysis has emerged as a prominent method for detecting prohibited substances because it offers numerous advantages over traditional matrices such as blood and urine in doping tests (e.g., the availability of historical evidence of drug use, ease of sample transportation and favorable sample storage conditions (at room temperature)) [20–22]. In fact, we have also successfully extended the detection times of substances such as GW1516 (peroxisome proliferator activated receptor δ agonist) and three hypoxia-inducible factor stabilizers (IOX4, vadadustat and daprodustat) from several weeks in plasma and urine [12,14,15,23,24] to at least 6 months [25–28]. To further extend the detection times of osilodrostat, the feasibility of hair analysis in horse manes is currently being investigated; the results will be presented in another scientific article in the near future.

Furthermore, anti-doping laboratories are increasingly exploring the feasibility of biomarkers for doping control and relevant studies to discover such biomarkers have been enthusiastically conducted worldwide [29–42]. Biomarkers are generally defined as indicators of prohibited substances identified by the up- or down-regulated levels of endogenous substances (e.g., metabolites, proteins, transcripts, or their combinations) through the administration of drugs [30,33,34]. Because the administration of osilodrostat inhibits the biosynthesis of cortisol, followed by compensatory stimulation of ACTH triggered by the decreased levels of endogenous cortisol in humans [1], we expected the biosynthesis of endogenous steroids in horses [39,43] to be facilitated by the administration of osilodrostat as well. In light of the pharmacological mechanism, we are conducting an ongoing related study of targeted steroidomics involved in the compensatory stimulation of ACTH to discover potential biomarkers for the administration of a CYP11B inhibitor (i.e., osilodrostat). Notably, if potential biomarkers specific to the administration of osilodrostat are discovered and established, the biomarker(s) might be applied to the detection of the same class of substances. The discovery of such biomarkers would enable the comprehensive and effective detection of prohibited substances (anabolic agents in this case), including newly emerged and truly unknown compounds, on the basis of their pharmacological effects in a more practical and feasible approach than direct detection of individual drugs and their metabolites. However, as we reported previously [44,45], the levels of endogenous substances in horses can vary significantly because of numerous confounding factors such as age, gender, feed, exercise, injury, temperature, climate, season and any other changes in physicochemical conditions. Therefore, thorough and comprehensive research will be required to establish biomarkers under various predictable conditions.

5. Conclusion

We developed and validated a quantification method for osilodrostat in horse urine equipped with a hydrolysis step using LC/ESI–HRMS and applied it to post-administration urine samples from four thoroughbreds (a gelding and three mares) to establish their elimination profiles. The elimination rates of osilodrostat in urine were very similar among them, indicating small gender differences. The enzymatical hydrolysis of phase II metabolites of osilodrostat and its mono-hydroxylated metabolite (M1c) effectively prolonged detection windows, resulting in successful detections of both osilodrostat and M1c in all of the collected samples for as long as two weeks post-administration. Given the availability of the reference material of osilodrostat, confirmatory analysis of osilodrostat would be the most appropriate for the doping control. To further extend the detection window of osilodrostat, we will investigate the feasibility of hair testing of osilodrostat in the near future.

Acknowledgments

We would like to thank A Nishizawa for technical assistance with the administration study at the Horse Racing School.

Funding Statement

This study was supported by a Grant-in-Aid from the Japan Racing Association (2024).

Financial disclosure

This study was supported by a Grant-in-Aid from the Japan Racing Association (2024). The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Competing interests disclosure

The authors have no competing interests or relevant affiliations with any organization or entity with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, stock ownership or options and expert testimony.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Ethical conduct of research

All experimental procedures were approved by the Animal Welfare and Ethics Committee of the Equine Research Institute of the Japan Racing Association (authorization number 23-31) and by the Laboratory of Racing Chemistry (authorization number 23-04).

References

Papers of special note have been highlighted as: • of interest; •• of considerable interest

  • 1.European Medicines Agency . Assessment report Isturisa [Internet]. Amsterdam: [updated 2019 November 14; cited 2024 May 15]. Available from: https://www.ema.europa.eu/en/documents/assessment-report/isturisa-epar-public-assessment-report_.pdf [Google Scholar]; •• An assessment report of osilodrostat, including detailed information on the pharmacological mechanism of osilodrostat and a pharmacokinetic study in humans.
  • 2.Duggan S. Osilodrostat: first approval. Drugs. 2020;80(5):495–500. doi: 10.1007/s40265-020-01277-0 [DOI] [PubMed] [Google Scholar]
  • 3.Thevis M, Kuuranne T, Geyer H. Annual banned-substance review-Analytical approaches in human sports drug testing 2021/2022. Drug Test Anal. 2023;15(1):5–26. doi: 10.1002/dta.3408 [DOI] [PubMed] [Google Scholar]
  • 4.International Federation of Horseracing Authorities . International Agreement on Breeding, Racing And Wagering And Appendixes (2023 edition with updated signatories) [Internet]. Paris: [updated 2024 May; cited 2024 May 15]. Available from: https://www.ifhaonline.org/resources/ifAgreement.pdf [Google Scholar]; • Describes the general requirements in horseracing, including those for prohibited substances in horses, as stipulated by the IFHA.
  • 5.Fédération Équestre Internationale . 2024 Equine Prohibited Substances List [Internet]. Lausanne: [updated 2023 December 4; cited 2024 May 15]. Available from: https://inside.fei.org/sites/default/files/2024%20Prohibited%20Substances%20List.pdf [Google Scholar]; • Describes a list of prohibited substances and medications in equestrian sports stipulated by the FEI.
  • 6.Armani S, Ting L, Sauter N, et al. Drug interaction potential of osilodrostat (LCI699) based on its effect on the pharmacokinetics of probe drugs of cytochrome P450 enzymes in healthy adults. Clin Drug Investig. 2017;37(5):465–472. doi: 10.1007/s40261-017-0497-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Balakirouchenane D, Vasseur A, Bonnet-Serrano F, et al. LC-MS/MS method for simultaneous quantification of osilodrostat and metyrapone in human plasma from patients treated for Cushing's Syndrome. J Pharm Biomed Anal. 2023;228:115316. doi: 10.1016/j.jpba.2023.115316 [DOI] [PubMed] [Google Scholar]
  • 8.Li L, Vashisht K, Boisclair J, et al. Osilodrostat (LCI699), a potent 11beta-hydroxylase inhibitor, administered in combination with the multireceptor-targeted somatostatin analog pasireotide: a 13-week study in rats. Toxicol Appl Pharmacol. 2015;286(3):224–233. doi: 10.1016/j.taap.2015.05.004 [DOI] [PubMed] [Google Scholar]
  • 9.Li W, Luo S, Rebello S, et al. A semi-automated LC-MS/MS method for the determination of LCI699, a steroid 11beta-hydroxylase inhibitor, in human plasma. J Chromatogr B Analyt Technol Biomed Life Sci. 2014;960:182–193. doi: 10.1016/j.jchromb.2014.04.012 [DOI] [PubMed] [Google Scholar]
  • 10.Ishii H, Ishikawa Y, Mizobe F, et al. Pharmacokinetic study of osilodrostat and identification of mono-hydroxylated metabolite in equine plasma for the purpose of doping control. Rapid Commun Mass Spectrom. 2024;38(5):e9695. doi: 10.1002/rcm.9695 [DOI] [PubMed] [Google Scholar]; •• Describes a pharmacokinetic study of osilodrostat in horse plasma.
  • 11.Ishii H, Shigematsu R, Takemoto S, et al. Metabolic pathway of osilodrostat in equine urine established through high-resolution mass spectrometric characterization for doping control. Crurrent Drug Metabolism. 2024. [DOI] [PubMed] [Google Scholar]; •• Describes the establishment of a metabolic pathway of osilodrostat in equine urine.
  • 12.Ishii H, Shibuya M, Kusano K, et al. Generic approach for the discovery of drug metabolites in horses based on data-dependent acquisition by liquid chromatography high-resolution mass spectrometry and its applications to pharmacokinetic study of daprodustat. Anal Bioanal Chem. 2022;414(28):8125–8142. doi: 10.1007/s00216-022-04347-2 [DOI] [PubMed] [Google Scholar]
  • 13.Ishii H, Leung GN, Yamashita S, et al. Comprehensive metabolic study of nicotine in equine plasma and urine using liquid chromatography/high-resolution mass spectrometry for the identification of unique biomarkers for doping control. J Chromatogr B Analyt Technol Biomed Life Sci. 2022;1190:123100. doi: 10.1016/j.jchromb.2022.123100 [DOI] [PubMed] [Google Scholar]
  • 14.Ishii H, Shibuya M, Kusano K, et al. Pharmacokinetic study of vadadustat and high-resolution mass spectrometric characterization of its novel metabolites in equines for the purpose of doping control. Crurrent Drug Metabolism. 2022;23(10):850–865. doi: 10.2174/1389200223666220825093945 [DOI] [PubMed] [Google Scholar]
  • 15.Ishii H, Shibuya M, So YM, et al. Comprehensive metabolic study of IOX4 in equine urine and plasma using liquid chromatography/electrospray ionization Q Exactive high-resolution mass spectrometer for the purpose of doping control. Drug Test Anal. 2022;14(2):233–251. doi: 10.1002/dta.3172 [DOI] [PubMed] [Google Scholar]
  • 16.Association of Official Racing Chemists . AORC Guidelines for the Minimum Criteria for Identification by Chromatography and Mass Spectrometry [Internet]. Storrs (CT): [updated 2024 April 9; cited 2024 May 15]. Available from: https://www.aorc-online.org/documents/aorc-ms-criteria-ilacg7/ [Google Scholar]; • Describes the minimum criteria for identifying prohibited substances (confirmatory analysis) using the techniques of chromatography and mass spectrometry for horseracing laboratories.
  • 17.Ishii H, Shimada M, Yamaguchi H, et al. A simultaneous determination method for 5-fluorouracil and its metabolites in human plasma with linear range adjusted by in-source collision-induced dissociation using hydrophilic interaction liquid chromatography-electrospray ionization-tandem mass spectrometry. Biomed Chromatogr. 2016;30(11):1882–1886. doi: 10.1002/bmc.3743 [DOI] [PubMed] [Google Scholar]
  • 18.Ishii H, Yamaguchi H, Mano N. Shifting the linear range in electrospray ionization by in-source collision-induced dissociation. Chem Pharm Bull (Tokyo). 2016;64(4):356–359. doi: 10.1248/cpb.c15-00741 [DOI] [PubMed] [Google Scholar]
  • 19.Ishii H, Yamaguchi H, Mano N. Expanding the versatility of a quantitative determination range adjustment technique using in-source CID in LC/MS/MS. Chromatography. 2017;38(2):59–63. doi: 10.15583/jpchrom.2017.004 [DOI] [Google Scholar]
  • 20.Thevis M, Geyer H, Tretzel L, et al. Sports drug testing using complementary matrices: advantages and limitations. J Pharm Biomed Anal. 2016;130:220–230. doi: 10.1016/j.jpba.2016.03.055 [DOI] [PubMed] [Google Scholar]
  • 21.Kintz P. Hair Analysis in Forensic Toxicology: An Updated Review with a Special Focus on Pitfalls. Curr Pharm Des. 2017;23(36):5480–5486. doi: 10.2174/1381612823666170929155628 [DOI] [PubMed] [Google Scholar]
  • 22.Kintz P, Gheddar L, Ameline A, et al. Hair testing for doping agents. What is known and what remains to do. Drug Test Anal. 2020;12(3):316–322. doi: 10.1002/dta.2766 [DOI] [PubMed] [Google Scholar]
  • 23.Ishii H, Leung GN, Yamashita S, et al. Doping control analysis of GW1516 in equine plasma using liquid chromatography/electrospray ionization Q-Exactive high-resolution mass spectrometry. Rapid Commun Mass Spectrom. 2020;34(23):e8920. doi: 10.1002/rcm.8920 [DOI] [PubMed] [Google Scholar]
  • 24.Ishii H, Shibuya M, Leung GN, et al. Metabolic study of GW1516 in equine urine using liquid chromatography/electrospray ionization Q-Exactive high-resolution mass spectrometry for doping control. Rapid Commun Mass Spectrom. 2021;35(5):e9028. doi: 10.1002/rcm.9028 [DOI] [PubMed] [Google Scholar]
  • 25.Ishii H, Shibuya M, Leung GN, et al. Detection and longitudinal distribution of GW1516 and its metabolites in equine hair for doping control using liquid chromatography/high-resolution mass spectrometry. Rapid Commun Mass Spectrom. 2021;35(8):e9050. doi: 10.1002/rcm.9050 [DOI] [PubMed] [Google Scholar]
  • 26.Ishii H, Shibuya M, So YM, et al. Long-term monitoring of IOX4 in horse hair and its longitudinal distribution with segmental analysis using liquid chromatography/electrospray ionization Q Exactive high-resolution mass spectrometry for the purpose of doping control. Drug Test Anal. 2022;14(7):1244–1254. doi: 10.1002/dta.3247 [DOI] [PubMed] [Google Scholar]
  • 27.Ishii H, Shibuya M, Kusano K, et al. First evidence of the incorporation of daprodustat and other hypoxia-inducible factor stabilizers into equine hair by passive transfer based on segmental quantitative analysis. J Pharm Biomed Anal. 2023;235:115600. doi: 10.1016/j.jpba.2023.115600 [DOI] [PubMed] [Google Scholar]
  • 28.Ishii H, Shibuya M, Kusano K, et al. Segmental analysis and long-term monitoring of vadadustat in equine hair for the purpose of doping control. J Anal Toxicol. 2023;47(7):623–631. doi: 10.1093/jat/bkad057 [DOI] [PubMed] [Google Scholar]
  • 29.Chan GH, Ho EN, Leung DK, et al. Targeted metabolomics approach to detect the misuse of steroidal aromatase inhibitors in equine sports by biomarker profiling. Anal Chem. 2016;88(1):764–772. doi: 10.1021/acs.analchem.5b03165 [DOI] [PubMed] [Google Scholar]
  • 30.Cawley AT, Keledjian J. Intelligence-based anti-doping from an equine biological passport. Drug Test Anal. 2017;9(9):1441–1447. doi: 10.1002/dta.2180 [DOI] [PubMed] [Google Scholar]; • Describes the concept of a biological passport for the indirect detection of the use of prohibited substances.
  • 31.Dahlgren AR, Knych HK, Arthur RM, et al. Transcriptomic markers of recombinant human erythropoietin micro-dosing in thoroughbred horses. Genes (Basel). 2021;12(12):1874. doi: 10.3390/genes12121874 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Loup B, Andre F, Avignon J, et al. miRNAs detection in equine plasma by quantitative polymerase chain reaction for doping control: assessment of blood sampling and study of eca-miR-144 as potential erythropoiesis stimulating agent biomarker. Drug Test Anal. 2022;14(5):953–962. doi: 10.1002/dta.3047 [DOI] [PubMed] [Google Scholar]
  • 33.Cawley A, Keen B, Tou K, et al. Biomarker ratios. Drug Test Anal. 2022;14(5):983–990. doi: 10.1002/dta.3250 [DOI] [PubMed] [Google Scholar]
  • 34.Cawley A. Biomarker analysis. Drug Test Anal. 2022;14(5):791–793. doi: 10.1002/dta.3268 [DOI] [PubMed] [Google Scholar]
  • 35.Cloteau C, Dervilly G, Kaabia Z, et al. From a non-targeted metabolomics approach to a targeted biomarkers strategy to highlight testosterone abuse in equine. Illustration of a methodological transfer between platforms and laboratories. Drug Test Anal. 2022;14(5):864–878. doi: 10.1002/dta.3221 [DOI] [PubMed] [Google Scholar]
  • 36.Tou K, Cawley A, Bowen C, et al. Towards non-targeted screening of lipid biomarkers for improved equine anti-doping. Molecules. 2022;28(1):312. doi: 10.3390/molecules28010312 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Tou K, Cawley A, Bowen C, et al. Measurements of hydrocortisone and cortisone for longitudinal profiling of equine plasma by liquid chromatography-tandem mass spectrometry. Drug Test Anal. 2022;14(5):943–952. doi: 10.1002/dta.3244 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Cloteau C, Dervilly G, Loup B, et al. Performance assessment of an equine metabolomics model for screening a range of anabolic agents. Metabolomics. 2023;19(4):38. doi: 10.1007/s11306-023-01985-0 [DOI] [PubMed] [Google Scholar]
  • 39.Viljanto M, Gray B, Scarth J. Presence and detection of endogenous steroids in the horse-A review. Drug Test Anal. 2023;1–22. doi: 10.1002/dta.3624 [DOI] [PubMed] [Google Scholar]
  • 40.Kikuchi M, Ishige T, Minamijima Y, et al. Identification of potential miRNA biomarkers to detect hydrocortisone administration in horses. Int J Mol Sci. 2023;24(19):14515. doi: 10.3390/ijms241914515 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Ishii H, Leung GN, Yamashita S, et al. Identification of potential biomarkers in urine and plasma after consumption of tobacco product in horses. Drug Test Anal. 2022;14(5):902–914. doi: 10.1002/dta.3242 [DOI] [PubMed] [Google Scholar]
  • 42.Ishii H, Shibuya M, Leung GN, et al. Additional studies on nicotine exposure in horses: accurate quantification and elimination profiles of potential biomarkers in plasma and urine. Rapid Commun Mass Spectrom. 2022;36(23):e9396. doi: 10.1002/rcm.9396 [DOI] [PubMed] [Google Scholar]
  • 43.Dirikolu L, Lehner AF. A review of current chemistry, pharmacology, and regulation of endogenous anabolic steroids testosterone, boldenone, and nandrolone in horses. J Vet Pharmacol Ther. 2023;46(4):201–217. doi: 10.1111/jvp.13397 [DOI] [PubMed] [Google Scholar]
  • 44.Ueda T, Tozaki T, Nozawa S, et al. Identification of metabolomic changes in horse plasma after racing by liquid chromatography-high resolution mass spectrometry as a strategy for doping testing. J Equine Sci. 2019;30(3):55–61. doi: 10.1294/jes.30.55 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Ohnuma K, Uchida T, Leung GN, et al. Establishment of a post-race biomarkers database and application of pathway analysis to identify potential biomarkers in post-race equine plasma. Drug Test Anal. 2022;14(5):915–928. doi: 10.1002/dta.3041 [DOI] [PubMed] [Google Scholar]

Articles from Bioanalysis are provided here courtesy of Taylor & Francis

RESOURCES