Validation of Activity-Based Screening for Synthetic Cannabinoid Receptor Agonists in a Large Set of Serum Samples

We previously reported an alternative screening method for the detection of synthetic cannabinoid receptor agonists (SCRAs)1 in serum and plasma samples (1). SCRAs are one of the largest groups of new psychoactive substances (NPS) worldwide. The increased availability and use of SCRAs (and other NPS) present substantial analytical and clinical challenges owing to the ever-increasing number of substances, their chemical diversity, and the morbidity/mortality associated with their use. Current analytical approaches mostly make use of targeted structure-based techniques, such as immunoassays or mass spectrometry-based methods. However, these have important limitations, including a lack of cross-reactivity and the need for prior knowledge of molecular identity. Moreover, the high potency of many SCRAs requires analytically sensitive detection methods, as low nanogram per milliliter concentrations in biological matrices are not uncommon. These “loopholes” in the current detection techniques underlie the recent research on “nontargeted,” activity-based screening methods for SCRAs and other NPS including synthetic opioids (1–4). The activity-based screening assays are based on the mechanism of receptor activation combined with the principle of functional complementation of a split Nanoluc® luciferase. Activation of a cannabinoid receptor (CB1/CB2), fused to one part of the luciferase, leads to recruitment of an engineered β-arrestin 2, fused to the other part. The resulting restoration of luciferase activity leads to measurable bioluminescence. This way, the bioassays allow detection of SCRAs purely based on their cannabinoid activity, requiring no prior knowledge about their structure. The main objective of this study was to validate the applicability and accuracy of activity-based cannabinoid bioassays as a first-line screening tool for the detection of SCRAs. A total of 471 serum samples, from patients with acute drug-related toxicity presenting to an Emergency Department in central London, UK, from April to December 2016, were analyzed (blind-coded) for the presence of SCRAs using the CB1 and CB2 bioassays (1). These results were compared to the drug(s) reported to have been taken by the patient and the bioanalytical data acquired through liquid chromatography coupled to high-resolution tandem mass spectrometry analysis (LC-HRMS/MS) (5). Ethical approval to perform this study was granted via the National UK Ethics approval reference 14/YH/1293. For 395 samples, enough sample volume (300–500 μL) was available to follow the sample preparation protocol previously described (1). The cannabinoid bioassays were able to identify all 52 SCRA-positive samples, yielding an analytical sensitivity of 100%, surpassing the 82% value previously reported in our proof-of-concept, in which 18 out of 22 SCRA-positives in a batch of 45 serum samples were correctly identified (1). In both this and our previous study, the detected SCRAs were present in serum samples in concentrations ranging from sub- to hundreds of nanogram per milliliter. The SCRAs involved in this study were (metabolites of) MDMB-CHMICA, AMB-CHMICA, cumyl-5F-PINACA, 5F-ADB, 5F-AKB-48, AB-CHMINACA, AMB- or EMB-CHMINACA, and AB-, AMB-, or EMB-FUBINACA. Of the 343 samples negative for SCRAs by LC-HRMS/MS analysis, 330 (96.2%) were correctly scored negative with both the CB1 and CB2 bioassays and 13 (3.8%) scored weakly positive for CB1 receptor activation only. In 6 of these 13 samples, tetrahydrocannabinol was detected, and thus, these samples should not be considered false positives because tetrahydrocannabinol is a partial CB agonist (1). There were therefore 7 false positives giving an analytical specificity of 97.9% (330/337), slightly lower than the 100% in our previous smaller study; however, this is still very high for a broad screening assay (1). Next to the 395 samples that had a volume ≥300 μL, an additional 76 serum samples were analyzed for which only 100 μL was available for sample preparation. Not completely unexpected, reducing the sample volume lowered the analytical sensitivity and specificity, to 88.9% (16/18) and 96.6% (56/58), respectively. Interestingly, in 2 false-positive cases, SCRA intake was claimed by the patient despite a lack of bioanalytical confirmation. Reevaluation of the bioanalytical data against a future updated library may retrospectively uncover the presence of SCRAs in these patients. It may also be possible that a combination of low-concentration SCRAs, not detectable via bioanalytical techniques, may have exerted an additive effect in these cases. In the absence of bioanalytical confirmation, and despite the concordance with the patients' claims, these samples should still be considered false positives. In conclusion, we confirmed the potential of activity-based screening for SCRAs via the successful application on a large set of authentic serum samples. The results obtained in this study demonstrate the potential for activity-based screening to complement conventional analytical methods in tackling the challenges associated with the detection of SCRAs (and other NPS) by serving as an alternative first-line “untargeted” screening method for these compounds. synthetic cannabinoid receptor agonist new psychoactive substance cannabinoid receptor liquid chromatography coupled to high-resolution tandem mass spectrometry analysis. We acknowledge the clinicians at Guy's and St Thomas' NHS Foundation Trust who recruited patients, collected samples and provided additional support for this study (Dr. John Archer, Dr. Matthew Blundell and Dr. Joanna White).