MSACL 2019 EU Abstract

Automated Sample Pretreatment for the Analysis of Synthetic Cannabinoids in Urine with LC-MS/MS

Stefan Eckelt (1), Carsten Mang (2)
(1) LABOR BERLIN, Berlin, Germany, (2) Hamilton Bonaduz AG, Bonaduz, Switzerland

Warning: Undefined variable $headshot in /var/www/html/view_abstract/view_abstract_in_program.php on line 704

Stefan Eckelt (Presenter) Labor Berlin >> POSTER (PDF)

Presenter Bio: Stefan Eckelt M.Sc. is working as scientific stuff member in the laboratory medicine and toxicology group of LABOR BERLIN.

Relevant Financial Disclosures (within past 24 months) Conflict mitigation NOT REQUIRED.

Introduction:
LC-MS/MS is a reliable, fast and highly sensitive technique used for multi analyte quantification of synthetic cannabinoids (SC). The analysis of the SC: JWH 018-M-N-5-OH, JWH 018-M-N-5-acid, JWH 073-M4-OH, JWH 073-M-N-B acid, JWH 081-5-OH, JWH 081-5-acid, JWH 122-5-OH, JWH 210-5-acid, JWH 250-5-OH, JWH 250-5-acid, AB-PINACA-OH, AB-PINACA-acid, AKB-48-OH, AKB-48-acid and PB-22-OH is performed for abstinence testing. Today’s laboratories are challenged with delivering high standards of laboratory services with fewer resources, so full workflow automation is key to delivering highly reliable results in the required timeframe. In particular, sample pretreatment is a bottleneck requiring several tedious pipetting steps that can lead to errors. This poster describes in detail sample pretreatment automation for the analysis of synthetic cannabinoids using the HAMILTON STARlet sample pretreatment robot combined with a Sciex QTRAP 3200.
Results:
The application was successfully developed as a walk away method; loading and unloading are the only manual steps. The automated workflow can handle up to 70 barcoded samples per run without the need of a worklist. The cooling module allows the samples to be stored in the instrument for at least 72 hours after processing and before analysis (decrease: 1.3-9.0%). It could be shown that the automated sample pretreatment with the HAMILTON STARlet performs very well with acceptable bias (0.5-14.7%) and precision (6.6-14.1%). The results correlated to those obtained with manual processing (R²: 0.7968-0.9195). The reduced correlation coefficients are due to the use of Protein LoBind plates compared to silanized vials in manual sample pretreatment. The synthetic cannabinoids have the property to bind rather well to plastic surfaces, however the LoBind plates have the best comparability to manual sample pretreatment. The results from conventional deep well plates differed more than 70%.
Conclusion:
With our automated method we show that sample pretreatment can be immediately implemented without the need for additional personnel.