Introduction: Drug checking services for harm reduction and overdose prevention have been implemented in many jurisdictions as a public health intervention in response to the opioid overdose crisis. This study demonstrates the first on‐site use of paper spray mass spectrometry for quantitative drug checking to address the limitations of current on‐site drug testing technologies.Methods: Paper spray mass spectrometry was used to provide on‐site drug checking services at a supervised consumption site in the Downtown Eastside of Vancouver, British Columbia, Canada during a 2‐day pilot test in August 2019. The method included the targeted quantitative measurement of 49 drugs and an untargeted full scan to assist in identifying unknown/unexpected components.Results: During the pilot, 113 samples were submitted for analysis, with 88 (78%) containing the client expected substance. Fentanyl was detected in 45 of 59 expected fentanyl samples, and in 50 (44%) samples overall at a median concentration of 3.6% (w/w%). The synthetic precursor of fentanyl, 4‐anilino‐N‐phenethyl‐piperidine (4‐ANPP), was found in 74.0% of all fentanyl samples at a median concentration of 2.2%, suggesting widespread poor manufacturing practices. Etizolam was detected in 10 submitted samples anticipated to be fentanyl at a median concentration of 2.5%. No clients submitting these samples expected etizolam or a benzodiazepine in their sample. In three instances, it was co‐measured with fentanyl, and in seven cases it was detected alone.Discussion and Conclusions: The quantitative capabilities and low detection limits demonstrated by paper spray mass spectrometry offer distinct benefits over existing on‐site drug checking methods and harm reduction services.
A direct mass spectrometry method utilizing reactive paper spray ionization was developed for sensitive cannabinoid quantitation in biofluid matrices. The ca. 2-minute sample measurements used on-paper derivatization to significantly increase paper spray mass spectrometry (PS-MS) positive ion mode sensitivity while minimizing sample preparation steps. Calibrations demonstrate high linearity, with R2 > 0.99 for (−)-trans-Δ9-tetrahydrocannabinol (THC) in oral fluid and 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THC-COOH) in urine. The limit of detection and lower limit of quantitation were 0.78 and 10 ng mL−1 for THC in oral fluid and 1.3 and 10 ng mL−1 for THC-COOH in urine, respectively. THC-COOH levels measured by reactive PS-MS in seven spiked human urine samples showed bias of −9.4 to 5.9%, and percent difference values of −16.8 to 9.8% in comparison with a reference LC-MS method. Based upon the method simplicity, validation experiments, sensitivity, and rapidity, we conclude that reactive PS-MS has potential applications for rapid cannabinoid drug testing in urine and oral fluid.