Immunoassay screening in urine for synthetic ... - Ruma Marker Test

Clin Chem Lab Med 2017; aop

Florian Franz, Verena Angerer, Hanna Jechle, Melanie Pegoro, Harald Ertl, Georg Weinfurtner, David Janele, Christian Schlögl, Matthias Friedl, Stefan Gerl, Reinhard Mielke, Ralf Zehnle, Matthias Wagner, Bjoern Moosmann and Volker Auwärter*

Immunoassay screening in urine for synthetic cannabinoids – an evaluation of the diagnostic efficiency DOI 10.1515/cclm-2016-0831 Received September 14, 2016; accepted December 12, 2016

Abstract Background: The abuse of synthetic cannabinoids (SCs) as presumed legal alternative to cannabis poses a great risk to public health. For economic reasons many laboratories use immunoassays (IAs) to screen for these substances in urine. However, the structural diversity and high potency of these designer drugs places high demands on IAs regarding cross-reactivity of the antibodies used and detection limits. *Corresponding author: Prof. Dr. Volker Auwärter, Institute of Forensic Medicine, Forensic Toxicology, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Albertstraße 9, 79104 Freiburg, Germany, Phone: + 49 761 203 6862, Fax: + 49 761 203 6826, E-mail: [email protected] Florian Franz and Verena Angerer: Institute of Forensic Medicine, Forensic Toxicology, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany; and Hermann Staudinger Graduate School, University of Freiburg, Freiburg, Germany Hanna Jechle, Melanie Pegoro and Bjoern Moosmann: Institute of Forensic Medicine, Forensic Toxicology, Medical Center – University of Freiburg, Faculty of Medicine, University of Freiburg, Freiburg, Germany Harald Ertl: Department of Toxicology and Drug Monitoring, Labor Lademannbogen, Hamburg, Germany Georg Weinfurtner: Clinical Chemistry Laboratory, medbo® – District Hospital for Mental Health, Regensburg, Germany David Janele: Clinic for Forensic Psychiatry, medbo® – District Hospital for Mental Health, Regensburg, Germany Christian Schlögl: Clinic for Forensic Psychiatry, medbo® – District Hospital for Mental Health, Parsberg, Germany Matthias Friedl: Department of Forensic Psychiatry, Klinikum am Europakanal, Erlangen, Germany Stefan Gerl: Department for Forensic Psychiatry, kbo-Inn-SalzachKlinikum, Wasserburg am Inn, Germany Reinhard Mielke: Forensic Psychiatry and Psychotherapy, ZfP Reichenau, Reichenau, Germany Ralf Zehnle: Forensic Psychiatry and Psychotherapy, ZfP Emmedingen, Emmendingen, Germany Matthias Wagner: Forensic Psychiatry and Psychotherapy, ZfP Calw, Calw-Hirsau, Germany

Methods: Two retrospective studies were carried out in order to evaluate the capability of two homogenous enzyme IAs for the detection of currently prevalent SCs in authentic urine samples. Urine samples were analyzed utilizing a ‘JWH-018’ kit and a ‘UR-144’ kit. The IA results were confirmed by an up-to-date liquid chromatography-electrospray ionization-tandem mass spectrometry (LC-ESI-MS/MS) screening method covering metabolites of 45 SCs. Results: The first study (n = 549) showed an 8% prevalence of SCs use (LC-MS/MS analysis) among inpatients of forensic-psychiatric clinics, whereas all samples were tested negative by the IAs. In a second study (n = 200) the combined application of both IAs led to a sensitivity of 2% and a diagnostic accuracy of 51% when applying the recommended IA cut-offs. Overall, 10 different currently prevalent SCs were detected in this population. The results can be explained by an insufficient cross-reactivity of the antibodies towards current SCs in combination with relatively high detection limits of the IAs. Conclusions: In light of the presented study data it is strongly recommended not to rely on the evaluated IA tests for SCs in clinical or forensic settings. For IA kits of other providers similar results can be expected. Keywords: abstinence control; drug screening; immunoassay; LC-MS/MS; synthetic cannabinoids; urine analysis.

Introduction Cannabinoid receptor agonists (‘synthetic cannabinoids’) (SCs) are a rapidly growing class of compounds within the group of so-called ‘new psychoactive substances’ (NPS) and are easily available via the Internet in the form of ready-to-smoke herbal mixtures or as ‘research chemicals’. These designer drugs are of high interest in clinical and forensic case work as they are often considered as legal alternatives to cannabis and increasingly consumed by young people [1–5].

Brought to you by | De Gruyter / TCS Authenticated Download Date | 2/1/17 11:45 AM

2 Franz et al.: Screening for synthetic cannabinoids in urine In contrast to cannabis, many SCs show an extremely high potency and act as full agonists at the CB1 receptor of the endocannabinoid system. Therefore, they can cause severe, sometimes life-threatening or even fatal intoxications [1, 6–8]. Especially with regard to their frequent and unpredictable composition, content of active ingredients, and inhomogeneity these ‘legal high’ products pose a great risk to public health [9]. While the prevalence of SC use is rather low in the general population when compared to cannabis use, consumption of SCs seems to be particularly popular among young men and experienced poly-drug users. Another group of consumers are people undergoing regular drug testing, such as professional athletes or soldiers [1, 3, 10–17]. It can be assumed that non-detectability of these substances in standard drug tests builds a strong motivation for inpatients of drug rehabilitation or forensicpsychiatric clinics and prisoners for SC use. Next to the aim of evaluating the diagnostic efficiency of commercially available Immunoassay/immunochemical assays (IAs) often used in clinical laboratories, we aimed to collect data on SC use among inpatients of forensic-psychiatric clinics in the south of Germany to verify this assumption. Typical issues requiring reliable analytical methods for detection of an SC uptake are drug abstinence control in the context of assessment of fitness to drive, court sanctions or withdrawal therapy, emergency intoxication cases, and postmortem toxicology. Failing to detect abuse of this new class of drugs may have serious consequences for individuals and society. For the past few years, different vendors offer automatable IAs for SCs promising a cost-effective highthroughput screening. Some of these assays have been scientifically evaluated in recent years [18–26]. The results of these studies showed that the kits offered were capable of detecting a broad spectrum of SCs and their respective metabolites. Although it was mentioned in some articles [20, 22–24, 26] that cross-reactivity with newer generation SCs has to be critically examined, the articles suggest that these tests can be used for a reliable screening approach. Consequently, IAs for SCs are widely utilized in clinical laboratories for routine analysis [10, 18]. From this perspective, it is of high interest to evaluate the capability of commonly used IAs to detect newer generation SCs.

Materials and methods Chemicals and reagents Formic acid (Rotipuran® ≥ 98%, p.a.), potassium hydrogen phosphate ( ≥ 99%, p.a.) and 2-propanol (Rotisolv® ≥ 99.95%, LC-MS

grade) were obtained from Carl Roth (Karlsruhe, Germany). Acetonitrile (LC-MS grade) (ACN), ammonium formate 10 M (99.995%) and potassium hydroxide (puriss. p.a. ≥ 86% (T) pellets) were from Sigma-Aldrich (Steinheim, Germany). All reference standards and deuterated standards were purchased from Cayman Chemical (Ann Arbor, MI, USA) and β-gucuronidase (expressed in Escherichia coli K12) from Roche Diagnostics (Mannheim, Germany). The utilized IA kits ‘Synthetic Cannabinoids-1’ (‘SC-1’) and ‘Synthetic Cannabinoids-2’ (‘SC-2’) as well as the solutions for calibration and control samples were from Immunalysis (Pomona, CA, USA). Deionized water was prepared using a Medica® Pro deionizer from ELGA (Celle, Germany). Blank urine samples were supplied by volunteers and tested for absence of SC metabolites prior to use. Mobile phase A (0.2% HCOOH and 2 mmol/L NH4+HCOO− in water) was freshly prepared prior to LC-MS/MS analysis. Pure ACN was used as mobile phase B.

Authentic urine samples Authentic urine samples were obtained in the context of drug screening for SCs. All analyses were conducted for diagnostic and therapeutic purposes and according to the client’s inquests. Hence, the study did not require approval by an Ethics Committee.

Study setting A The scope of this substudy was to assess the prevalence of SC abuse within a population undergoing regular drug screening without preselection. A second aspect was to evaluate the diagnostic efficiency of IA screenings for SCs applied in routine analysis of clinical chemistry laboratories. Seven forensic-psychiatric clinics located in the German federal states of Baden-Wuerttemberg (BaWu) (n = 3) and Bavaria (n = 4) participated in the study. All patients (Table 1) of the respective wards were required to supply a urine sample on a fixed date without preannouncement. Sampling was performed under direct supervision of a staff member and the samples were divided into two aliquots directly afterwards. One aliquot was used for the immunochemical screening in the clinical chemistry laboratory of the medbo® in Regensburg, Germany, the other aliquot was sent to the Institute of Forensic Medicine in Freiburg, Germany, for LC-MS/ MS confirmatory analysis. After analysis the results of both methods were compared. In total, urine samples of 549 patients were collected from October to November 2014.

Study setting B The second study was conducted to evaluate the performance parameters of both IAs under realistic conditions in cooperation with an independent laboratory (Labor Lademannbogen, Hamburg, Germany) using the same IAs as the laboratory of medbo® Regensburg for routine SC screening. For this purpose urine samples screened routinely for SC metabolites by LC-MS/MS analysis at the Institute of Forensic Medicine in Freiburg were collected from January to June 2015. Out of this collective 100 negative samples and 100 consecutive samples positive for metabolites of a single SC were included in the study


Franz et al.: Screening for synthetic cannabinoids in urine 3 Table 1: Results of the screening for synthetic cannabinoids in the individual clinics by two immunochemical assays and LC-MS/MS. Number of samples of each hospital, sex and age distribution of the inpatients are given. Hospital BaWu A BaWu Ba BaWu Ca Bavaria Ab Bavaria Bb Bavaria Cb Bavaria Db Total a

Samples 75 66 49 126 135 45 53 549

93% 100% 96% 94% 94% 100% 100% 96%

7% 0% 4% 6% 6% 0% 0% 4%

Age [a] (Øc; md)

IA positive

LC-MS/MS positive

21–63 (35; 34) 20–74 (35; 33) 18–59 (32; 29) 17–54 (28; 25) 20–68 (33; 31) 23–50 (33; 32) 22–57 (34; 33) 17–74 (32; 30)

0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%

0.0% 3.0% 14.3% 24.6% 0.7% 2.2% 0.0% 7.7%

Clinic in the federal state of Baden-Wuerttemberg. bClinic in the federal state of Bavaria. cAverage age. dMedian age.

a

Table 2: Screening results for synthetic cannabinoids (SCs) by two immunoassays within a collective of 100 negative and 100 positive (for metabolites of a single SC) urine samples. Samples were collected consecutively and confirmed by LC-MS/MS analysis. Samples

Sex

Age [a] (Øa; mb)

LC-MS/MS analysis Substances

Positive

Negative Total

97 (97%) 3 (3%)

14–49 (29; 27)

Total

90 (90%) 10 (10%)

16–69 (32; 30)

Total

187 (94%) 13 (7%)

14–69 (30; 29)

AB-CHMINACA AB-FUBINACA ADB-CHMINACA AM-2201 MDMB-CHMICA 5F-PB-22 THJ-018 False positive

IA analysis d

#

‘SC-2’e

100

2

1

27 4 6 1 59 1 2 0

0 0 0 0 0 0 2 0

0 0 0 0 0 0 0 1

100

198

199

200

200

200

c

‘SC-1’

Average age. bMedian age. cNumber of samples. d‘Synthetic Cannabinoids-1’ kit, manufacturer recommended cut-off. e‘Synthetic Cannabinoids-2’ kit, manufacturer recommended cut-off.

a

without any further preselection (Table 2). These urine samples were blinded and re-analyzed using the two IAs. Samples positive for metabolites of more than one SC were excluded to allow evaluation of the IAs’ cross-reactivities with regard to single SCs.

Immunochemical screening All urine samples of setting A were analyzed after centrifugation on a Cobas Integra® 800 analyzer (Roche Diagnostics, Mannheim, Germany) utilizing the ready-to-use homogeneous enzyme immunoassays (HEIA™) ‘SC-1’ and ‘SC-2’ (Immunalysis, Pomona, CA, USA). In case of the IA ‘SC-1’ calibration was performed with a blank sample and a calibration solution consisting of JWH-018 pentanoic acid (20 ng/mL). Additionally, a low (10 ng/mL) and high (30 ng/mL) JWH018 pentanoic acid control sample was analyzed. A cut-off of 20 ng/ mL as recommended by the manufacturer was applied. Calibration for the IA ‘SC-2’ was performed with a blank sample and a calibration

solution consisting of UR-144 pentanoic acid (10 ng/mL). The control samples contained 5 and 15 ng/mL UR-144 pentanoic acid, respectively, and a cut-off of 10 ng/mL as recommended by the manufacturer was applied. The cross-reactivities of the respective antibodies in the two kits as stated by the manufacturer are listed in Supplemental Table 1 [27]. Urine samples of setting B were analyzed on a Cobas Integra® 400 Analyzer (Roche Diagnostics, Mannheim, Germany) under identical conditions. SPSS® Statistics ver. 22.0 (IBM, Ehningen, Germany) software was used for generation of receiver operating characteristic (ROC) plots and evaluation of different cut-off values.

LC-MS/MS screening Sample preparation was performed by adding 0.5 mL phosphate buffer (pH 6) and 30 μL β-glucuronidase to 0.5 mL of urine and incubation at 45 °C for 1 h. Afterwards, 1.5 mL ACN fortified with the internal standards (2 ng/mL solution of d4-JWH-018 pentanoic acid,


4 Franz et al.: Screening for synthetic cannabinoids in urine d5-AM-2201 4-OH-pentyl, d5-JWH-018 4-OH-pentyl, d5-JWH-018 5-OHpentyl, d5-JWH-073 3-OH-butyl, d5-JWH-073 4-OH-butyl, d5-JWH-073 butanoic acid, d5-JWH-122 5-OH-pentyl, d5-JWH-200, d5-JWH-250, d5-JWH-250 4-OH-pentyl, d5-JWH-250 5-OH-pentyl, d7-JWH-015, d7-JWH-018 6-OH-indole, d7-JWH-073 6-OH-indole, and d9-JWH-081) and 0.5 mL of a 10 M ammonium formate solution was added. The mixture was shaken, centrifuged, 1 mL of the ACN phase transferred into a separate vial and evaporated to dryness under a stream of nitrogen. Finally, the samples were reconstituted in 200 μL mobile phase A/B (50/50, v/v) prior to LC-MS/MS analysis. Furthermore, one positive control consisting of 5 μL reference standard solution (1 μg/mL) and 10 μL of an ACN extract of pooled authentic SC positive urine samples (containing metabolites of AB-CHMINACA, AB-FUBINACA, AB-PINACA, ADB-CHMINACA, APICA, MDMB-CHMICA, STS-135, THJ018 and THJ-2201) as well as negative control samples (blank urine samples) were prepared accordingly. The LC-MS/MS system consisted of a Dionex UltiMate® 3000RS UHPLC (Thermo Scientific, Dreieich, Germany) coupled to an API 5000™ triple quadrupole instrument with a TurboIonSpray® interface (Sciex, Darmstadt, Germany). Separation was performed on a Luna® C18(2) column (100 Å, 150 × 2 mm, 5 μm) with a corresponding guard column (4 × 2 mm) (both Phenomenex, Aschaffenburg, Germany) applying gradient elution as follows: Starting condition of mobile phase B was 50%, linearly increased to 60% within 3.5 min, further increased to 70% within 2.0 min, further increased to 75% within 1.0 min, further increased to 80% within 1.0 min, further increased to 90% in 1.5 min, held for 2.5 min, decreased to starting conditions of 50% in 0.5 min and held for 3.0 min for re-equilibration. The flow rate was set to 0.25 mL/min and the autosampler as well as the column oven temperature to 10 °C and 40 °C, respectively. The injection volume was 10 μL. A scheduled Multiple Reaction Monitoring™ (sMRM) method in positive electrospray ionization mode consisting of at least two characteristic ion transitions of metabolites of 45 different SCs was used for qualitative sample analysis. The main phase I metabolites of the following SCs were covered: 5F-AB-PINACA, 5F-AKB-48, 5F-AMB, 5F-PB22, AB-001, AB-CHMINACA, AB-FUBINACA, AB-PINACA, ADBICA, ADB-CHMINACA, ADB-FUBINACA, ADB-PINACA, AKB-48, AM-2201, AM-694, AMB, APICA (SDB-001), BB-22, EAM-2201, FUB-PB-22, JWH007, JWH-018, JWH-019, JWH-072, JWH-073, JWH-081, JWH-122, JWH200, JWH-203, JWH-210, JWH-250, JWH-307, JWH-398, MAM-2201, MDMB-CHMICA, NNEI, PB-22, RCS-4, SDB-006, STS-135, THJ-018, THJ-2201, UR-144, UR-144 Isomer, XLR-11. The ion transitions of ADBCHMINACA metabolites were only included for the samples analyzed within substudy B (ADB-CHMINACA was not available on the market when the samples for substudy A were collected). Declustering potentials (DP), entrance potentials (EP), collision energies (CE) and cell exit potentials (CXP) were carefully optimized for each analyte. For some SC metabolites reference standards were not commercially available. In these cases, the optimized mass spectrometric parameters of the respective parent compound were assigned, assuming a similar fragmentation behavior of a parent compound and its phase I metabolites. For detailed information on the MS parameters of the covered metabolites see Supplemental Table 2. Ion source temperature and ion source voltage were set to 600 °C and + 4500 V, respectively. Curtain gas (N2) pressure as well as ion source gas 1 and 2 (compressed air) pressures were set to 40 psi. Collision gas (N2) pressure was set 6 psi. The LC-MS/MS screening method utilized for quantification was fully validated for 31 SC metabolites of 14 SCs according to the guidelines of the German Society of Toxicological and Forensic Chemistry

(GTFCh) [28]. The validation parameters are shown in Supplemental Tables 3 and 4. Briefly, limits of detection (LODs) and lower limits of quantification (LLOQs) ranged from 0.01 ng/mL to 0.5 ng/mL and 0.05 ng/mL to 0.5 ng/mL, respectively. Linearity was given for all analytes from their respective LLOQs to 10 ng/mL. The intraday and interday precisions for each compound were below 12% (RSD ≤ 12%) at all quality control concentration levels with a bias of ≤ 6.9% (accuracy). Autosampler stability was given for 21.5 h. Degradation was observed to be 0.5) and of a random distribution (dotted line, AUC = 0.5) are given for comparison (right).

Discussion Prevalence Considering the results from the individual clinics (Table 1) it becomes clear that the prevalence rate strongly varies from institution to institution (0% – 25%, mean 7.7%) and will certainly also vary over time. No apparent difference could be observed regarding the positivity rate between the two respective federal states (BaWu and Bavaria). Nevertheless, SC use among inpatients of forensic-psychiatric clinics is common and may hamper the therapeutic process. Similar prevalence rates can be assumed among other populations undergoing frequent drug abstinence control. Therefore, rigorous screening for SC abuse is an essential part of proving actual drug abstinence.

Cross-reactivity The data collected within setting B (Table 2) demonstrate that the antibodies used in both IA kits do not show sufficient cross-reactivity towards the metabolites of newer generation SCs, in particular to the valine- or tert-leucine derivative type compounds like AB-CHMINACA, AB-FUBINACA, ADB-CHMINACA, and MDMB-CHMICA. This is not surprising recalling the distinct differences between the chemical structures of these compounds and JWH-018 or UR-144 (Figure 1). The same issue

applies for metabolites of AB-PINACA, APICA and 5F-PB22 which were not detected by the IAs in urine samples analyzed within setting A. In contrast, the ‘SC-1’ kit seems to be suitable for the detection of THJ-018 metabolites which can be explained by the close similarity of its chemical structure to the structure of JWH-018. Moreover, even in the case of structural similarity, crossreactivity of the antibody with the main metabolites, with metabolic pathways showing substance-specific and interindividual variations, is not guaranteed and only a small number of metabolites are available as pure standards for cross-reactivity testing (see Supplemental Table 1). Considering the analytical results of over 3500 forensic serum samples screened for SCs from January 2012 till June 2015 (overall rates of positive samples 19%) it can be observed how the spectrum of compounds actually consumed shifted over the last 3 years (analyses performed at the Institute of Forensic Medicine Freiburg, Germany) (Figure 3). This dynamic in the availability of new compounds poses a great challenge for clinical and forensic laboratories. In the case of IAs, fundamental structural changes in the compounds would require development of further antibodies specifically binding to the new structures. On the one hand development of a new assay is time consuming and causes delay, and on the other hand with a growing number of assays the cost-efficiency becomes questionable in comparison to LC-MS/MS methodology. Kronstrand et al. published an evaluation of the ‘SC-1’ kit in comparison to an LC-Q-ToF-MS based screening approach in 2014 [26]. Their Brought to you by | De Gruyter / TCS Authenticated Download Date | 2/1/17 11:45 AM

Franz et al.: Screening for synthetic cannabinoids in urine 7

Figure 3: Prevalence of consumed synthetic cannabinoids (SCs) visualized by analyzed authentic serum samples between January 2012 and June 2015. The number of samples tested positive for a respective SC was compared to the number of serum samples tested positive for any SC within the same quarterly period (maximum 76%). Q: quarterly period.

results showed an additional cross-reactivity of the IA for MAM-2201 metabolites and comparable performance parameters for both techniques by analyzing a collective of urine samples from April 2013. These samples contained metabolites of AM-2201, JWH-018, JWH-073, JWH-122, JWH210, MAM-2201, and UR-144, all of them showing acceptable cross-reactivities for the antibody except metabolites of UR-144. Nevertheless, Kronstrand et al. concluded that an LC-MS/MS screening approach is a ‘superior strategy to IAs’ due to the rapid change of SCs in ‘legal high’ products. The results of our study clearly confirm this conclusion by analyzing samples collected only 18 months later (substudy A). Figure 3 reflects the market changes within this time period. Another promising approach based on cannabinoid receptor interaction has been published recently by Cannaert et al. [29]. In this work an assay was introduced that could solve the cross-reactivity issue as long as the SC metabolites present in urine show sufficient activity at the cannabinoid receptors and might be an interesting alternative for immunochemical drug testing in the future.

Detection limits Even authentic urine samples containing metabolites of AM-2201 (setting B) and JWH-122 (setting A) remained

undetected by both IAs, although the cross-reactivities assessed by the manufacturer for the ‘SC-1’ kit seem promising (133% for AM-2201 6-hydroxyindole metabolite and 25% for JWH-122). Considering these cross-reactivities (see Supplemental Table 1), a cut-off of 20 ng/mL probably leads to detection limits in the same range at best. In comparison, the measured concentrations of AM-2201 metabolites in the authentic urine sample were 0.05 ng/mL JWH-018 5-OH-pentyl, 0.06 ng/mL JWH-018 pentanoic acid, and 0.05 ng/mL JWH-072 propionic acid. Concentrations of metabolites in the JWH-122 positive urine sample were approximately 0.02 ng/mL for JWH-122 4-OH-pentyl and approximately 0.003 ng/mL in the case of the 6-OH-indole metabolite (concentrations extrapolated from the lowest calibrator). These examples demonstrate the limited applicability of the IAs for abstinence control. Considering the average concentrations of SC metabolites detected in forensic urine samples (own unpublished data, Table 3) these constraints become even more obvious. Evaluation of the data from over 5700 forensic urine samples (years 2013 and 2014) show that the detected metabolite concentrations were below 10 ng/mL in 77% – 100% of the cases when referring to the single analytes (analysis performed at the Institute of Forensic Medicine Freiburg, Germany). Data published earlier also confirmed this finding [30]. Consequently, even detection Brought to you by | De Gruyter / TCS Authenticated Download Date | 2/1/17 11:45 AM

8 Franz et al.: Screening for synthetic cannabinoids in urine Table 3: Concentration range of selected analytes in authentic urine samples tested positive for synthetic cannabinoids (years 2013 and 2014). Analyte

JWH-018 4-OH-indole JWH-018 5-OH-indole JWH-018 4-OH-pentyl JWH-018 5-OH-pentyl JWH-018 pentanoic acid JWH-073 3-OH-butyl JWH-073 4-OH-butyl JWH-073 butanoic acid JWH-122 5-OH-indole JWH-122 4-OH-pentyl JWH-122 5-OH-pentyl JWH-122 pentanoic acid JWH-210 5-OH-indole JWH-210 4-OH-pentyl JWH-210 5-OH-pentyl JWH-210 pentanoic acid UR-144 4-OH-pentyl UR-144 5-OH-pentyl UR-144 pentanoic acid XLR-11 4-OH-pentyl

< 0.05, ng/mLa

0.05–10, ng/mL

> 10, ng/mLb

Samples, nc

n.a. 20% 18% 16% 13% 10% 38% 16% 78% 18% 23% 26% 20% 100% 14% 22% 10% 19% 6% 0%

n.a. 80% 60% 68% 65% 90% 58% 71% 22% 75% 70% 68% 50% 0% 86% 78% 76% 76% 72% 86%

n.a. 0% 21% 15% 22% 0% 4% 13% 0% 8% 7% 5% 0% 0% 0% 0% 14% 5% 23% 14%

n.d. 10 121 243 223 10 24 55 9 40 97 38 2 1 28 9 21 42 53 7

Concentration below the lowest calibrator. bConcentration above the highest calibrator. cNumber of samples tested positive for the respective metabolite. n.a., not applicable. n.d., not detected.

a

of metabolites of ‘first generation’ SCs would lead to falsenegative results in the majority of the cases when applying the recommended cut-off.

Limitations The applied LC-MS/MS assay was not fully validated for all analytes covered. As Gerostamoulos et al. recently outlined, a full validation of up-to-date multi-target analyte screening methods in the field of NPS is often not possible due to non-availability of commercial reference standards [31]. However, for the target compounds showing promising cross-reactivity that were used to assess the immunoassays’ sensitivity, full validation data are provided. The result of an immunoassay test always reflects the binding of the antibody to compounds of the whole metabolic spectrum present in a urine sample with different binding activities for each metabolite including unspecific binding. Therefore, the signal intensity is not necessarily expected to correlate to the added signal intensities of the main metabolites monitored by the LC-MS/MS confirmation method. The applied IA cut-offs

(20 ng/mL and 10 ng/mL, respectively) might lead to positive results even in cases where lower concentrations were measured a pplying the LC-MS/MS method (see Table 3). However, this is only the case if cross-reactivity can be expected at all (e.g. close structural similarity of the analytes). The prevalence of SC use among inpatients of forensic-psychiatric clinics is not representative in a wider sense. The analyzed population of 549 patients was a representative sample for the seven included hospitals at that time since all forensic-psychiatric patients of these hospitals supplied a urine sample. However, the results only refer to a rather small geographic region in southern Germany and may not be generalized. For the evaluation of the diagnostic efficiency of both IAs only a limited number of positive samples were analyzed. This was a result of collecting authentic urine samples which had to be positive for metabolites of only a single SC. Nevertheless, the study evaluates the performance of the tests under realistic conditions without bias due to preselection. A drawback of this design is that we did neither cover predefined ranges of analyte concentrations, nor achieve a uniform distribution of the detected SCs (the results reflect the prevalence of the active ingredients).

Conclusions On the basis of the obviously insufficient cross-reactivity towards the currently prevalent compounds and their metabolites as well as due to the high cut-off values applied, the utility of the tested IAs has to be strongly questioned. The high proportion of false-negative results may lead to diagnostic errors and faulty treatment. Therefore, the authors would advise against the use of such IAs in both clinical and forensic settings, unless sufficient cross-reactivity to the compounds available on the ‘legal high’ market can be guaranteed and the detection limits of the tests are improved. Furthermore, similar results can be expected for IAs of other providers as long as they are based on similar antibodies and test principles. At present, the only valid and reliable alternative for detection of SC use is a carefully designed and frequently updated LC-MS/ MS based method. Acknowledgments: The authors would like to thank Raija Treffeisen, Kathrin Riedy and Manuel Sandmeyer for technical assistance.


Franz et al.: Screening for synthetic cannabinoids in urine 9

Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission. Research funding: The research activities of the Institute of Forensic Medicine Freiburg were financially supported by the ‘Prevention of and Fight against Crime’ program of the European Commission Directorate-General for Justice (JUST/2013/ISEC/DRUGS/AG/6421) and the Deutsche Forschungsgemeinschaft (GZ: INST 380/92-1 FUGG). Employment or leadership: None declared. Honorarium: None declared. Competing interests: The funding organization(s) played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.

References 1. Castaneto MS, Gorelick DA, Desrosiers NA, Hartman RL, Pirard S, Huestis MA. Synthetic cannabinoids: epidemiology, pharmacodynamics, and clinical implications. Drug Alcohol Depend 2014;144:12–41. 2. Kemp AM, Clark MS, Dobbs T, Galli R, Sherman J, Cox R. Top 10 facts you need to know about synthetic cannabinoids: not so nice spice. Am J Med 2016;129:240–4. 3. Bretteville-Jensen AL, Tuv SS, Bilgrei OR, Fjeld B, Bachs L. Synthetic cannabinoids and cathinones: prevalence and markets. Forensic Sci Rev 2013;25:7–26. 4. EMCDDA. European drug report 2016: trends and developments. Lisbon: Publications Office of the European Union, 2016. 5. UNODC. World drug report 2015. United Nations Office on Drugs and Crime, 2015. 6. Chung H, Lee J, Kim E. Trends of novel psychoactive substances (NPSs) and their fatal cases. Forensic Toxicol 2016;34:1–11. 7. Hermanns-Clausen M, Kneisel S, Szabo B, Auwärter V. Acute toxicity due to the confirmed consumption of synthetic cannabinoids: clinical and laboratory findings. Addiction 2013;108:534–44. 8. EMCDDA. Early warning system reports. Accesible via EDND database of the EMCDDA. Accessed: 11 Feb 2014. 9. Moosmann B, Angerer V, Auwärter V. Inhomogeneities in herbal mixtures: a serious risk for consumers. Forensic Toxicol 2015;33:54–60. 10. Castaneto MS, Scheidweiler KB, Gandhi A, Wohlfarth A, Klette KL, Martin TM, et al. Quantitative urine confirmatory testing for synthetic cannabinoids in randomly collected urine specimens. Drug Test Anal 2015;7:483–93. 11. Wohlfarth A, Scheidweiler KB, Castaneto M, Gandhi AS, Desrosiers NA, Klette KL, et al. Urinary prevalence, metabolite detection rates, temporal patterns and evaluation of suitable LC-MS/MS targets to document synthetic cannabinoid intake in US military urine specimens. Clin Chem Lab Med 2015;53:423–34. 12. Heltsley R, Shelby MK, Crouch DJ, Black DL, Robert TA, Marshall L, et al. Prevalence of synthetic cannabinoids in US athletes: initial findings. J Anal Toxicol 2012;36:588–93.

13. Palamar JJ, Acosta P. Synthetic cannabinoid use in a nationally representative sample of US high school seniors. Drug Alcohol Depend 2015;149:194–202. 14. Spaderna M, Addy PH, D’Souza DC. Spicing things up: synthetic cannabinoids. Psychopharmacology 2013;228:525–40. 15. Bonar EE, Ashrafioun L, Ilgen MA. Synthetic cannabinoid use among patients in residential substance use disorder treatment: prevalence, motives, and correlates. Drug Alcohol Depend 2014;143:268–71. 16. Loeffler G, Delaney E, Hann M. International trends in spice use: prevalence, motivation for use, relationship to other substances, and perception of use and safety for synthetic cannabinoids. Brain Res Bull 2016;126:8–26. 17. Loeffler G, Hurst D, Penn A, Yung K. Spice, bath salts, and the US military: the emergence of synthetic cannabinoid receptor agonists and cathinones in the US armed forces. Mil Med 2012;177:1041–8. 18. Arntson A, Ofsa B, Lancaster D, Simon JR, McMullin M, Logan B. Validation of a novel immunoassay for the detection of synthetic cannabinoids and metabolites in urine specimens. J Anal Toxicol 2013;37:284–90. 19. Barnes AJ, Spinelli E, Young S, Martin TM, Kleete KL, Huestis MA. Validation of an ELISA synthetic cannabinoids urine assay. Ther Drug Monit 2015;37:661–9. 20. Barnes AJ, Young S, Spinelli E, Martin TM, Klette KL, Huestis MA. Evaluation of a homogenous enzyme immunoassay for the detection of synthetic cannabinoids in urine. Forensic Sci Int 2014;241:27–34. 21. Castaneto MS, Desrosiers NA, Ellefsen K, Anizan S, Martin TM, Klette KL, et al. Method validation of the biochip array technology for synthetic cannabinoids detection in urine. Bioanalysis 2014;6:2919–30. 22. Mohr AL, Ofsa B, Keil AM, Simon JR, McMullin M, Logan BK. Enzyme-linked immunosorbent assay (ELISA) for the detection of use of the synthetic cannabinoid agonists UR-144 and XLR-11 in human urine. J Anal Toxicol 2014;38:427–31. 23. Rodrigues WC, Catbagan P, Rana S, Wang GH, Moore C. Detection of synthetic cannabinoids in oral fluid using ELISA and LC-MS-MS. J Anal Toxicol 2013;37:526–33. 24. Spinelli E, Barnes AJ, Young S, Castaneto MS, Martin TM, Klette KL, et al. Performance characteristics of an ELISA screening assay for urinary synthetic cannabinoids. Drug Test Anal 2015;7:467–74. 25. Schwope DM, Milman G, Huestis MA. Validation of an enzyme immunoassay for detection and semiquantification of cannabinoids in oral fluid. Clin Chem 2010;56:1007–14. 26. Kronstrand R, Brinkhagen L, Birath-Karlsson C, Roman M, Josefsson M. LC-QTOF-MS as a superior strategy to immunoassay for the comprehensive analysis of synthetic cannabinoids in urine. Anal Bioanal Chem 2014;406:3599–609. 27. Immunalysis. K2-II-Synthetic-Cannbinoids-HEIA-Info-SheetVer.-A-Final [PDF]. Available at: http://immunalysis.com/ wp-content/uploads/2013/03/MKT-1051-K2-II-Synthetic-Cannabinoids-HEIA-Info-Sheet-Ver.-A-Final.pdf. Accessed: 17 Jun 2015. 28. Peters FT, Hartung M, Herbold M, Schmitt G, Daldrup T, Mußhoff F. Anhang B Anforderungen an die Validierung von Analysenmethoden. Toxichem Krimtech 2009;76:185–208. 29. Cannaert A, Storme J, Franz F, Auwärter V, Stove CP. Detection and activity profiling of synthetic cannabinoids and


10 Franz et al.: Screening for synthetic cannabinoids in urine metabolites with a newly developed bio-assay. Anal Chem 2016;88:11476–85. 30. Hutter M, Moosmann B, Kneisel S, Auwärter V. Characteristics of the designer drug and synthetic cannabinoid receptor agonist AM-2201 regarding its chemistry and metabolism. J Mass Spectrom 2013;48:885–94. 31. Gerostamoulos D, Elliott S, Walls HC, Peters FT, Lynch M, Drummer OH. To measure or not to measure? That is the NPS question. J Anal Toxicol 2016;40:318–20.

Supplemental Material: The online version of this article (DOI: 10.1515/cclm-2016-0831) offers supplementary material, available to authorized users. Article note: Parts of this work have been presented at the 19th GTFCh Symposium (Mosbach, Germany), the 53rd TIAFT Meeting (Florence, Italy), the 7th EAFS Conference (Prague, Czech Republic), the 94th DGRM Meeting (Leipzig, Germany) and the 12th DGKL Meeting (Leipzig, Germany).
