EVALUATION OF SEMI-TARGETED AND NON-TARGETED (‘DISCOVERY’) SCREENING TOOLS

Importance of the topic

Analytical toxicology faces a growing challenge with the rapid appearance of novel psychoactive substances. High resolution mass spectrometry, especially time-of-flight in MSE mode, provides comprehensive datasets that support both targeted identification of known drugs and discovery of unknown analytes.

Objectives and study overview

This study evaluates semi-targeted and non-targeted (discovery) screening workflows alongside traditional targeted analysis. Seventeen postmortem blood samples were analysed to compare performance in identifying over 1300 drug substances and metabolites.

Methodology and used instrumentation

• Samples: Seventeen postmortem blood specimens.
• Instrumentation: Waters ACQUITY UPLC I-Class, Xevo G2-XS QTOF mass spectrometer, UNIFI informatics platform.
• Data acquisition: TOF-MSE mode capturing simultaneous low-energy (precursor accurate mass) and high-energy (fragment ions) spectra.
• Data processing:
• Targeted analysis matched data to a library of over 1300 compounds using retention time, mass accuracy (±5 ppm), and diagnostic fragments.
• Semi-targeted analysis employed Molfile structures with in-silico fragmentation to generate theoretical fragments for matching.
• Non-targeted discovery workflow proposed elemental compositions, performed external library searches, applied in-silico fragmentation, and compared to high-energy data.

Main results and discussion

• Targeted analysis detected 138 drug-related substances; 92.7% were confirmed by at least one diagnostic fragment.
• Semi-targeted screening confirmed 88% of those detections via theoretical fragment matching and identified three additional substances.
• Discovery workflow independently verified 85% of the targeted identifications, demonstrating robust non-targeted performance.

Benefits and practical applications

• Automated semi-targeted screening increases throughput and confidence, even when reference standards are lacking.
• Discovery tools enable retrospective examination of datasets and identification of unknown compounds without prior target lists.
• Combined workflows streamline forensic toxicology procedures and improve overall detection capability.

Future trends and applications

Advancements in spectral library expansion, machine learning-driven fragmentation prediction, and cloud-based data integration will further enhance non-targeted screening. Emerging applications include real-time toxicological monitoring in clinical and environmental settings.

Conclusion

The integrated use of targeted, semi-targeted, and non-targeted screening with high resolution mass spectrometry and informatics significantly improves the detection and confirmation of known and emerging psychoactive substances, offering a flexible, efficient approach for forensic toxicology.

References

1. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA)
2. United Nations Office on Drugs and Crime (UNODC)
3. UK Home Office Forensic Early Warning System (FEWS)