Drug testing is essential to ensure fair competition and protect the well-being of the athletes and jockeys in horse racing. The current state-of-the-art technology used in equine drug testing is an LC-MS-based targeted approach. Despite its sensitivity and specificity, such an approach is unable to detect emerging doping agents beyond the target lists of drugs. To address this issue, we previously developed an analytical methodology for the comprehensive untargeted detection of unknown drugs in equine plasma. With this methodology, we performed the analysis of official post-race equine plasma samples to provide insights into emerging doping agents in such samples.
Aided by a Racing Medication and Testing Consortium grant, we conducted untargeted detection of emerging doping agents in post-race equine plasma samples collected at six racetracks in a horse racing jurisdiction over a period of more than a year. For such detection, different screening approaches were initially compared, and an appropriate one was chosen. The raw LC-MS data with high-resolution and high mass accuracy scans from routine targeted drug screening analysis were reused for the initial untargeted detection screening, and this screening led to inclusion lists for further LC-MS/MS reanalysis of the plasma samples. The acquired MS/MS data were utilized to detect (identify) emerging doping agents.
An R script was built to aid with the automated filtering of the screening and detection results in Compound Discover software, by the criteria of the ratio of mean (ROM) and outlier index (OLI). A Python script was developed to determine if detected (identified) substances were relevant to doping, via artificial intelligence. The results from this study will be presented.