Applying Non-Targeted Analysis to Food Safety Pesticide Screening

Significance of the topic

Residual pesticide monitoring is essential for food safety, consumer protection and regulatory compliance. Traditional targeted LC/TQ and GC/MS workflows deliver excellent sensitivity and quantitative precision but are limited to predefined analyte lists. High-resolution mass spectrometry (HRMS) combined with non-targeted analysis (NTA) enables simultaneous targeted quantitation and broad suspect screening in a single run, increasing chemical coverage, detecting metabolites and unexpected or newly regulated compounds, and supporting more proactive surveillance strategies in complex food matrices.

Objectives and overview of the study

The study aimed to develop and demonstrate a unified LC/QTOF workflow that supports both validated targeted quantitation and large-scale non-targeted screening. Specific goals were to evaluate the capability of NTA to detect a broad pesticide panel at regulatory-relevant concentrations and to apply the same workflow to unknown real-world samples to reveal additional residues, metabolites and unexpected contaminants outside routine target lists.

Methodology and instrumentation

• Sample preparation: Standard QuEChERS extraction for food commodities. A representative matrix test used olive oil spiked with pesticides at 10 µg/kg (typical EU reporting limit/MRL for many compounds).
• Injection strategy: A modified POISe (Pretreatment Optimised Injection Sequence) approach to improve retention and peak shape for polar analytes — 1 µL sample mixed with 10 µL water repeated four times for a total injection volume of 4 µL; autosampler held at 12 °C.
• Chromatography (Nexera X2): Shim-pack Velox Biphenyl column (2.1 × 100 mm, 2.7 µm) with biphenyl guard (2.1 × 5 mm, 2.7 µm) at 45 °C; flow rate 0.4 mL·min−1. Mobile phases: A = water + 2 mM ammonium formate + 0.004% formic acid; B = methanol + 2 mM ammonium formate + 0.004% formic acid. Gradient from 3.5% to 100% B over a 16.4 min cycle with specified ramps and rinses to minimize carryover.
• Mass spectrometry (LCMS-9050 QTOF): ESI positive mode, interface 300 °C, 1.5 kV; heat block 400 °C, desolvation line 250 °C; gas flows: nebulizing 2 L·min−1, heating 15 L·min−1, drying 5 L·min−1. TOF-MS ranged m/z 140–925. DIA-MS/MS acquisition covered m/z 40–1000 using 31 sequential windows (29 ms each), CE 5–55 V, 1 s cycle time to collect fragment spectra for library matching.
• Data processing: LabSolutions Insight Profiler for feature detection, alignment and batch screening. Processing used predefined parameters including non-targeted feature detection and a feature inclusion threshold (summed spectral intensity >10,000). Compound identification relied on an in-house curated HRMS pesticide MS/MS library (>500 compounds) and supplemental third-party libraries (e.g., Food Safety Mass Spectral Library from Wageningen University).

Main results and discussion

• Sensitivity and detection breadth: A complex test sample (olive oil) spiked with 280 pesticides at 10 µg/kg was analyzed. The NTA workflow detected and identified all 280 pesticide parent ions at this concentration, with MS/MS spectral library matches that met SANTE HRMS quantitative guidelines.
• Unknown sample screening: Application to real-world commodity samples (e.g., baby spinach, nectarine, grape, leek, dried bean, dry apricot) confirmed expected residues previously quantified by LC/TQ and revealed additional compounds outside routine target lists. Notable findings included repeated detection of acetamiprid in multiple samples (including high levels in baby spinach), pyrimethanil at high levels in nectarine, and a combination of boscalid, fluopyram, penconazole and pyrimethanil in grape.
• Metabolites and extended search space: Using third-party libraries, metabolites and transformation products such as acetamiprid-N-dimethyl and propamocarb N-oxide were identified, illustrating the value of broad libraries for capturing biotransformation products. Detection of dimethomorph — a fungicide not renewed by the EU in 2024 but still authorized in some jurisdictions — demonstrates the method's ability to flag regulatory-relevant residues outside prioritized target lists.
• Data quality and workflow automation: The LabSolutions Insight Profiler batch workflow automated feature detection, alignment and library matching, enabling high-throughput screening without bespoke per-sample parameterisation. A low-intensity feature threshold and MS/MS matching provided robust identifications at MRL-level concentrations.

Benefits and practical applications of the method

• Unified analysis: Combines validated targeted quantitation with wide-scope NTA in a single LC/QTOF run, reducing analysis time and method duplication.
• Expanded chemical surveillance: Enables detection of unexpected pesticides, metabolites and transformation products not included in routine panels — valuable for regulatory monitoring, incident investigation and post-market surveillance.
• Scalability and throughput: Automated batch processing with predefined settings supports routine implementation in high-throughput laboratories and QA/QC environments.
• Regulatory relevance: Demonstrated capability to detect residues at EU SANTE reporting limits (10 µg/kg) and to identify compounds of regulatory concern, aiding compliance and risk assessment.

Future trends and potential applications

• Library expansion and open resources: Growth of curated, open-access HRMS spectral libraries (including metabolites and transformation products) will improve identification confidence and broaden coverage.
• In-silico prediction and machine learning: Integration of predicted spectra and retention-time models can extend identification to compounds absent from empirical libraries and reduce false negatives.
• Quantitative workflows for metabolites: Developing validated quantitative methods for prioritized metabolites and transformation products will be increasingly important for regulatory compliance and exposure assessment.
• Interoperability and automation: Better integration with LIMS, automated reporting, and standardized confidence metrics will facilitate routine adoption in enforcement laboratories.
• Regulatory adoption of HRMS NTA: As confidence in HRMS-based workflows grows, regulators may increasingly accept HRMS data for screening and confirmatory purposes, prompting harmonized guidelines for NTA performance and reporting.

Conclusion

A unified LC/QTOF NTA workflow using a modified injection strategy, DIA acquisition and automated processing with LabSolutions Insight Profiler reliably detected and identified a broad pesticide panel (280 compounds) at 10 µg/kg in a complex matrix and revealed additional residues and metabolites in real samples. The approach offers a scalable, high-coverage alternative to multiple targeted methods, improving the ability of laboratories to detect known and unexpected agrochemical residues and to respond to evolving regulatory requirements.

Reference

1. Padilla-González F, Rizzo S, Dirks C, Bergkamp W, Rasker S, Aloisi I. Creation of an Open-Access High-Resolution Tandem Mass Spectral Library of 1000 Food Toxicants. Analytical Chemistry. 2025;97(43):23822–23830. DOI: 10.1021/acs.analchem.5c03020.
2. European Commission. Analytical quality control and method validation procedures for pesticides residues analysis in food and feed. SANTE/11312/2021 v2. 2021. (SANTE guidance on method validation and QC for pesticide residues.)