http://www.jfsmonline.com/article.asp?issn=2349-5014;year=2018;volume=4;issue=4;spage=184;epage=191;aulast=Anzillotti
Many new psychoactive substances (NPSs) with different chemical structures have emerged in the illicit drug market in the last decade. NPSs are different chemical compounds sold online through the e-commerce and the deep web as legal substitutes for classical drugs of abuse, including synthetic cannabinoids, synthetic cathinones, phenethylamines, piperazines, or substances not relating to any of these groups and plant-based materials.[1] The ease of NPS distribution favored their quick spreading worldwide through different channels. However, as soon as NPSs are scheduled, new derivatives appear on the market; therefore, the number of NPS reported by the European Monitoring Centre for Drugs and Drug Addiction increases each year.[2] This rapid increase of NPS sets new challenges not only in drug prevention and legislation but also in clinical and forensic toxicology, as the acute and chronic toxicity of many of these compounds is still partially unknown. Hence, the identification in biological samples is of great concern for forensic and clinical toxicologists, to evaluate the spread of NPS among population. According to the 2016 European Early Warning System Report, the largest substance categories monitored are the synthetic cannabinoids (over 160 substances, including 11 new cannabinoids reported in 2016), followed by the synthetic cathinones (over 100 substances, 14 reported for the first time in 2016).[3] Even within the same class (i.e., synthetic cannabinoids), as soon as legislation is passed banning their use, different compounds show up in the next wave. These substances show different function group chemistry that dictates a sample extraction procedure that will capture the various chemical functionalities. Some NPSs are extremely potent in terms of dosage, so that they may only be present at trace levels. Hence, the necessity and the ability to analyze the complex chromatographic data in the presence of large amounts of coextractant material. These materials are also structurally similar in terms of chromatographic retention time (RT) and mass spectral appearance. Data analyses need to be able to identify the subtle differences in these species and be able to detect such substances in complex mixtures.[3],[4]
Many analytical methods were developed for NPS determination in biological fluids, such as oral fluid,[4],[5],[6] blood,[7],[8],[9] or urine.[10],[11]To date from a recent search in literature, only few studies deal with the determination of NPS in hair;[12],[13],[14],[15],[16],[17] to the best of our knowledge, no paperwork mentioned the determination of these classes of substances together. The present pilot study was aimed at the development of a simple method in gas chromatography/mass spectrometry (GC/MS) for the determination of eight NPSs of different classes (mainly synthetic cannabinoids), the use of cannabinoids, and, at the same time, the evaluation of methadone therapy in hair matrix, within our routine analyses control for patients with methadone treatment or from autopsy cases. The development of the method involved an extraction technique based on incubation in concentrated sodium hydroxide (NaOH) solution, providing a dissolution of the keratin matrix.[12],[13] Hair sampling is easy to perform, not invasive, and relatively stable, moreover less affected by adulterants.[18] Hair samples allow a retrospective determination of the drug use history depending basically on hair length due to their accumulation in keratin, taking into account that head hair grows at an average rate of 1 cm circa each month,[19] and being able to confirm long-term exposure: it is, therefore, a reliable and valuable tool to assess chronic use of drugs in a specific population.[18] Moreover, parent drugs prevalently accumulate in hair and keratinized matrices in general with respect to unmetabolized drugs,[20] avoiding hydrolysis steps for the determination of metabolites.
Reagents and standards
Water, sodium dodecyl sulfate (SDS), acetone, acetonitrile, formic acid, phosphate buffer and methanol, chloroform and isopropanol, NaOH, hexane, and ethyl acetate were purchased from Sigma Aldrich, Milano, Italy. 2-Ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP), methadone (1,1-diphenyl-1-(2-dimethylaminopropyl)-2-butanone), cannabinol (6, 6, 9-trimethyl-3-pentyl-6H-dibenzo[b, d]pyran-1-ol), cannabidiol (2-[(1R,6R)-3-methyl-6-(prop-1-en-2-yl)cyclohex-2-en-1-yl]-5-pentylbenzene-1,3-diol), Δ9-tetrahydrocannabinol (THC), UR144 (1-pentyl-1H-indol-3-yl)(2, 2, 3, 3-tetramethylcyclopropyl)methanon), CP47497 (2-[(1R,3S)-3-hydroxycyclohexyl]-5-(2-methyl-2-octanyl)phenol) and its homolog CP47497-C7, 1-([5-fluoropentyl]-1H-indol-3-yl)-(naphthalen-1-yl)methanone (AM2201), (1-hexyl-1H-indol-3-yl)-1-naphthalenyl-methanone (JWH-019), (4-methoxy-1-naphthalenyl) (1-pentyl-1H-indol-3-yl)methanone (JWH-081), (4-methyl-1-naphthalenyl) (1-pentyl-1H-indol-3-yl)methanone (JWH-122), 1-(1-pentyl-1H-indol-3-yl)-2-(2-methoxyphenyl)-ethanone (JWH-250), tetrahydrocannabinol-D3 (THC-D3), EDDP-D3, and methadone-D3 were supplied from LGC standards (Milan, Italy). Standard compounds were stored according to supplier recommendations until their use.
Calibration and sample preparation
Hair strands were collected either from routine analyses or from autopsies, cut from the posterior vertex region of the head, close to the scalp since this region is associated with least variation in growth rates (the amount required by SOHT guidelines is a pencil thickness.[18]) Hair sample aliquots were washed with 3 mL × 3 of a solution of SDS 1%, rinsed twice with 3 mL of distilled water, and then twice with 1 mL of acetone. After drying, each sample was segmented in samples of 1 cm each circa, then each one shred and grinded into small pieces of 1 mm circa (30 mg in weight).
A working solution mixture of deuterated drugs of abuse (mix drugs-Deut) containing EDDP D3, methadone D3, and THC D3 at 10 μg/mL was prepared by proper dilution of the standard solutions and stored at −20°C until use.
Individual methanolic stock solutions were used to prepare a working solution at a concentration of 10 μg/mL. Calibration curves were prepared by addition of the appropriate amount of cannabinoids to 30 mg of blank hair sample (collected from three different drug-free subjects) to obtain the following concentrations: 0.1, 0.5, 1, 2, 5, 10, and 20 ng/mg.
Thirty milligrams of hair samples was put in a vial and 3 μL mix drugs-Deut plus 500 μL of NaOH were added to digest hair sample at 90°C for 30 min. Then, the sample was extracted with 1.5 mL of a mixture composed by hexane: ethyl acetate (9:1) by automated shaking for 10 min and centrifuged for 5 min at 2000 rpm. The supernatant was then transferred into another vial and evaporated under a gentle stream of nitrogen to dryness. After the evaporation step, the sample was reconstituted with 100 μL of MeOH and 2 μL was injected in the GC/MS equipment.
Gas chromatography/mass spectroscopy equipment
The GC instrument was an Agilent 7820A (Agilent Technologies, Santa Clara, CA, USA), and the parameters chosen for optimization were the following: the liner temperature was held at 270°C; helium was used as a carrier gas at a constant flow of 40 mL/min. The oven temperature started from 100°C, then by 20°C/min was held at 250°C for 10 min, then at 20°C/min to 280°C and was held for 5 min and finally to 320°C for 4 min.
The detection system was an Agilent 5977B single quadrupole MS operating in selective ion monitoring (SIM) mode. The column was a J and W DB-5 (5% phenylmethyl silicone) capillary column (30 mm × 0.25 mm. i.d., 0.25 μm film thickness, Agilent Technologies). Characteristic ion fragments of investigated compounds were chosen and optimized injecting the individual methanolic solutions in scan mode before developing the analytical method in SIM [as reported in [Table 1]a. After pretreatment of the sample, 2 μL was injected into the instrument.
Method validation
The method was validated according to the Food and Drug Administration guidelines[21] and was evaluated for linearity, limit of detection (LOD), limit of quantification (LOQ), lowest limit of quantitation (LLOQ), accuracy, and repeatability. The linearity of the assay was calculated by the method of least squares and expressed as coefficient of determination (R2). Calibration curves were prepared in triplicate in 3 different days by adding to blank hair samples a mixture of the commercially available standards at a concentration of 10 μg/mL in the proper amount to obtain the range of concentration and to determine the LOD and of quantification (LOQ). These parameters were studied using serial dilutions of the substances of interest in matrix in triplicate and analyzed in 5 different days. Repeatability and accuracy were assessed at three concentrations: low (quality control [QC] 1), medium (QC2), and high (QC3) injected in quintuplicate in 3 different days and were expressed, respectively, as CV% and E%. The parameters studied are listed below.
Specificity
Ten negative hair samples from voluntary subjects were collected and analyzed to determine specificity and verify, therefore, the absence of interfering peaks that could hinder the analytes. Specificity was also assessed by analyzing samples spiked with a sample at a concentration of 500 pg/mg of compounds with most common illicit or therapeutic drugs (such as cocaine and metabolites, opiates, benzodiazepines, and various antipsychotic drugs). Hence, satisfactory specificity was established if no interfering signals were found in terms of characteristic fragments and RT related to endogenous or exogenous compounds.
Limit of detection and limit of quantification
The LOD was calculated at a concentration value giving an S/N >3 for at least three ion fragments for each substance while the LOQ was considered the concentration value giving an S/N >10 for three ion fragments and acceptable accuracy and precision (%CV and %E <20%). LLOQ was calculated at the concentration value giving an S/N ratio >5. These parameters were studied using scalar dilutions of the substances of interest in hair in quintupled.
Linearity
The linearity of the method for each compound was studied in the range from the LLOQ of each substance to 20 ng/mg, performing triplicate analyses for each level. Calibration curves were built by linear regression of the area ratio of each substance with their internal standard (IS) versus the concentration of analyte.
Accuracy and precision
QC samples were prepared at three concentration levels: low (QC1), medium (QC2), and high concentrations (QC3). Accuracy and precision were assessed by analyzing the QCs in quintuplicate in three different days and were expressed respectively as %error (%E) and standard deviation (STD).
Memory effect
Memory effect, intended as carryover of analytes from sample to sample, was evaluated: two blank samples were injected after each run of spiked samples at 50 and 100 ng/mg and analyzed after positive samples.
Identification criteria
The criteria to be fulfilled for the identification of analytes were RT, the presence of three ion fragments, and their relative ion intensities. For the identification of an analyte, RT should not vary more than ±2.5%; relative ion intensities should not vary more than ±20% for ions with relative intensities >50%, ±25% for ions with relative intensities between 10% and 50%, and ±50% for ions with relative intensities <10%, with respect to a spiked control sample.
Recoveries
Recovery experiments were performed by comparing the analytical results for extracted samples at three concentrations (low 1 ng/mg, medium 10 ng/mg, and high 20 ng/mg) with samples spiked with standards after the extraction procedure that represent 100% recovery.
Specificity
The proposed method demonstrated its specificity for the detection and quantification of methadone and its main metabolite EDDP, including cannabinoids and the most common NPS in hair samples, verifying the absence of peaks that could interfere with the substances of interest.
Limit of detection and limit of quantification
All the analytes investigated were detectable in the range from 0.05 ng/mg to 0.5 ng/mg. The only exception was AM2201 which could be determined at 1 ng/mg [Table 1]b.
Linearity
From calibration curves, built by linear regression of the area ratio of each substance with their IS versus the concentration of analyte, the linearity of the assay was calculated by the method of least squares and expressed as coefficient of determination (R2). The method was linear in the range from LOQ to the highest concentration assessed with quadratic regression coefficients (R2) ranging from 0.9978 to 0.9997. R2 is reported for each analyzed substance in [Table 1]b.
Accuracy, precision, and recoveries
Accuracy and precision were expressed, respectively, as %CV and STD, and the results are shown in [Table 1]b. CV% values are lower than 20% for low concentrations and lower than 15% for high concentrations; therefore, according to the guidelines, the method showed acceptable accuracy and precision values. As expected after liquid/liquid extraction, a low matrix effect was observed: recovery percentages were very high (around 100%) for almost all the monitored compounds, except for JWH 019 and JWH 122, as shown in [Table 1]b.
In summary, validation parameters such as LODs, LOQs, repeatability, accuracy, and linearity were satisfactory for its application on real specimens. LODs, LOQs, R%CV, standard deviation, and the mean concentration for the analyzed compounds are reported in [Table 1]b (nominal values of QCs 1, 2, and 3 were 1 ng/mg, 5 ng/mg, and 20 ng/mg, respectively). Accuracy and repeatability were acceptable for all the analytes at their respective LOQs. Recovery experiments varied from 58.3% to 103.0%. [Figure 1] shows an extracted ion chromatogram of the quantifier ion for all the substances investigated.
Application to real samples
Since the recent application of the protocol in our laboratory, the described method was applied on 15 authentic specimens from our cases: five showed the presence of methadone and EDDP: for example, the first analyzed hair sample's segments were from a female subject found dead in her apartment from which we were able to collect 15 cm of hair strand. Nine segments (S) of 1 cm circa were prepared and analyzed and showed the following results: the root was positive for methadone at 4.31 ng/mg, 6.42 ng/mg (S-1), 5.47 ng/mg (S-2), 2.51 ng/mg (S-3), 53.19 ng/mg (S-4), 129.57 ng/mg (S-5), 361.08 ng/mg (S-6), 86.87 ng/mg (S-7), and 149.80 ng/mg (S-8), including its metabolite EDDP at 7.98 ng/mg (root), 7.22 ng/mg (S-1), 10.23 ng/mg (S-2), 7.65 ng/mg (S-3), 38.93 ng/mg (S-4), 72.38 ng/mg (S-5), 5 ng/mg (S-6) 23.47 ng/mg (S-7), and 118.47 ng/mg, respectively (S-8). In [Figure 2], chromatogram of authentic postmortem hair sample (S-5) positive for methadone and EDDP is shown at 129.57 ng/mg and 72.38 ng/mg, respectively. The subject had a well-known history of substance abuse and resulted positive to many xenobiotics in biological fluids as well. Results obtained demonstrated the proficiency of the developed method to determine, with a satisfactory sensitivity and sensibility, the drugs of abuse in hair samples involved in the study rapidly and with a simple sample pretreatment.
Although a few immunochemical rapid tests can detect few NPS, the gold standard for their detection is chromatography coupled to mass spectrometry. Drugs levels in hair are considerably lower than those found in matrices such as blood or urine; therefore, single or tandem mass spectrometry is employed for confirmation tests. Hair testing analysis provides a retrospective timeframe via segmental analysis, due to the larger detection window when compared with other specimens (up to months, depending on strands lengths); another advantage of hair sampling is low potential for donor manipulation. In the present work, we developed and validated an analytical method to employ either after autopsy or in a population of routine analysis for methadone treatment, showing a particularly high potential for the identification of NPS users. As our study confirms, the method is suitable for analyses of studied compounds; however, there are some limitations hereby discussed. The positivity rate obtained from the study is indeed influenced by the number of samples that we could be able to collect and analyze, although it has to be stressed that these are preliminary results and that more samplings and analyses will be performed; therefore, a significant greater number of authentic cases will be assayed.
In addition, since NPS comprehends a very wide range of substances with different chemical structures, it must be noticed that the application of our analytical procedure might not be suitable for other compounds (i.e., cathinones). In particular, heating the samples as a pretreatment procedure might affect and potentially have a destructive effect on the chemical stability of NPS with low boiling points or different structures, such as mephedrone or other synthetic derivatives.[22] For example, ester analogs (e.g., PB-22) decompose (or participates in the ester-exchange reactions) in the injection port; another example is that cyclopropyl urinary metabolites (e.g., UR-144) undergo a thermal degradation mainly in GC column.[23],[24],[25] To avoid such issues, when performing splitless injection, an injector temperature of 270°C and a surface deactivated injector liner without glass wool minimizes the degradation and enhances the sensitivity. These results indicate that special attention is required for GC-MS analysis of NPS.[23] Similar mass spectra are sometimes obtained by GC–MS analyses due to regio- and ring-substituted analogs available on the market: the misidentification of these analogs arises when comparing data only with the available mass spectra. When tandem and high-resolution MS are used to identify the conformational isomers or regioisomers, such misidentification does not occur. Moreover, compounds that are thermally unstable might decompose in the GC injection port, especially those with polar groups (i.e., amino or hydroxyl groups) that can cause a polar interaction with the column stationary phase, leading to poor detection of the analyte. To overcome these problems, derivatization step should be added onto the extraction method for these compounds; however, this step would be time-consuming and potentially toxic. From our validation study, satisfactory results on our set of analytes can be obtained overcoming the derivatization step.
Indeed, synthetic cannabinoids are not ideal compounds for GC-MS analysis because they are neutral to weakly acidic compounds and have high molecular weight. However, concerning the substances taken into consideration in the validation of the presented method, no destructive effect was noticed; therefore, the method is applicable only for the relevant compounds included in this study and more experiments will be performed when other NPS will be included in the procedure. The phenomenon might interest some metabolites that, however, are not the focus of our study, because the parent compounds are the main targets in hair analyses.
Hair analysis for NPS is still at an early stage of development, particularly on the toxicological screening side. The proposed method allows for the identification of synthetic cannabinoids, cannabinoids, and methadone (including its main metabolite) in hair samples with a simple sample pretreatment.
The identification of NPS in biological samples is one of the emerging challenges for forensic laboratories due to the necessity of detection and confirmation of a very large class of substances, often not structurally correlated. Due to the reoccurring threat of synthetic cannabinoids to public health and their rapidly increasing abuse worldwide, it is necessary to develop reliable analytical methods for their detection in different biological matrices. SCs are constantly being modified and rapidly becoming widely available; therefore, laboratories should update their scope for detecting the most prevalent compounds at specific times.
Blood and urine are the first choice of sample for testing; however, hair is often used as an alternative matrix in repeated drug exposure. The method validation presented herein is a straightforward, selective, and accurate method for the determination of some drugs belonging to the CP and aminoalkylindole structural classes. The described method can be useful not only in the forensic investigation of NPS-related addiction histories but also in epidemiological and retrospective studies on the spread of NPS among specific safety-sensitive social workers, such as drivers.
This study has been approved by the local ethics committee and the consent was exempted.
| ORIGINAL ARTICLE | |
|
Determination of methadone and eight new psychoactive substances in hair samples by gas chromatography/mass spectrometry
Luca Anzillotti, Luca Calò, Marianna Giacalone, Antonio Banchini, Rossana Cecchi
Department of Medicine and Surgery, Institute of Legal Medicine, University of Parma, Parma, Italy
| Date of Web Publication | 27-Dec-2018 |
     |
Correspondence Address:
Dr. Luca Anzillotti
Institute of Legal Medicine, University of Parma, Via A. Gramsci 14, Parma 43126
Italy
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/jfsm.jfsm_22_18
| Abstract |
Many new psychoactive substances (NPSs) with different chemical structures have emerged in the illicit drug market in the last decade. The present work was aimed at the development of a simple method in gas chromatography/mass spectrometry (MS) for the determination of NPS of different classes, the use of cannabinoids, and, at the same time, the evaluation of methadone therapy in hair matrix, within our routine analysis control for methadone treatment or from autopsy cases. The determination of synthetic cannabinoids and methadone therapy used an extraction method based on incubation in concentrated sodium hydroxide (NaOH) solution, providing a dissolution of the keratin matrix. The described method was applied on 15 authentic specimens from our cases: five showed the presence of methadone and 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP). The described method can be useful not only in the forensic investigation of NPS-related addiction histories but also in epidemiological and retrospective studies on the spread of NPS among specific safety-sensitive social workers. The GC instrument was an Agilent 7820A (Agilent Technologies, Santa Clara, CA, USA), and the detection system was an Agilent 5977B single quadrupole MS operating in selective ion monitoring mode. Validation parameters such as limit of detections (LODs), limit of quantifications (LOQs), repeatability, accuracy, and linearity were satisfactory for its application on real specimens. LODs, LOQs, R%CV, standard deviation, and the mean concentration for the analyzed compounds are reported in Table 1b. Accuracy and repeatability were acceptable for all the analytes at their respective LOQs. Recovery experiments varied from 58.3% to 103.0%, thus allowing the application on authentic specimens. The described method can be useful not only in the forensic investigation of NPS-related addiction histories but also in epidemiological and retrospective studies on the spread of NPS among specific safety-sensitive social workers, such as drivers.
Keywords: Drugs of abuse, forensic, gas chromatography/mass spectrometry, hair, new psychoactive substance, toxicology
| How to cite this article: Anzillotti L, Calò L, Giacalone M, Banchini A, Cecchi R. Determination of methadone and eight new psychoactive substances in hair samples by gas chromatography/mass spectrometry. J Forensic Sci Med 2018;4:184-91 |
| How to cite this URL: Anzillotti L, Calò L, Giacalone M, Banchini A, Cecchi R. Determination of methadone and eight new psychoactive substances in hair samples by gas chromatography/mass spectrometry. J Forensic Sci Med [serial online] 2018 [cited 2018 Dec 29];4:184-91. Available from: http://www.jfsmonline.com/text.asp?2018/4/4/184/248696 |
| Introduction |  |
Many new psychoactive substances (NPSs) with different chemical structures have emerged in the illicit drug market in the last decade. NPSs are different chemical compounds sold online through the e-commerce and the deep web as legal substitutes for classical drugs of abuse, including synthetic cannabinoids, synthetic cathinones, phenethylamines, piperazines, or substances not relating to any of these groups and plant-based materials.[1] The ease of NPS distribution favored their quick spreading worldwide through different channels. However, as soon as NPSs are scheduled, new derivatives appear on the market; therefore, the number of NPS reported by the European Monitoring Centre for Drugs and Drug Addiction increases each year.[2] This rapid increase of NPS sets new challenges not only in drug prevention and legislation but also in clinical and forensic toxicology, as the acute and chronic toxicity of many of these compounds is still partially unknown. Hence, the identification in biological samples is of great concern for forensic and clinical toxicologists, to evaluate the spread of NPS among population. According to the 2016 European Early Warning System Report, the largest substance categories monitored are the synthetic cannabinoids (over 160 substances, including 11 new cannabinoids reported in 2016), followed by the synthetic cathinones (over 100 substances, 14 reported for the first time in 2016).[3] Even within the same class (i.e., synthetic cannabinoids), as soon as legislation is passed banning their use, different compounds show up in the next wave. These substances show different function group chemistry that dictates a sample extraction procedure that will capture the various chemical functionalities. Some NPSs are extremely potent in terms of dosage, so that they may only be present at trace levels. Hence, the necessity and the ability to analyze the complex chromatographic data in the presence of large amounts of coextractant material. These materials are also structurally similar in terms of chromatographic retention time (RT) and mass spectral appearance. Data analyses need to be able to identify the subtle differences in these species and be able to detect such substances in complex mixtures.[3],[4]
Many analytical methods were developed for NPS determination in biological fluids, such as oral fluid,[4],[5],[6] blood,[7],[8],[9] or urine.[10],[11]To date from a recent search in literature, only few studies deal with the determination of NPS in hair;[12],[13],[14],[15],[16],[17] to the best of our knowledge, no paperwork mentioned the determination of these classes of substances together. The present pilot study was aimed at the development of a simple method in gas chromatography/mass spectrometry (GC/MS) for the determination of eight NPSs of different classes (mainly synthetic cannabinoids), the use of cannabinoids, and, at the same time, the evaluation of methadone therapy in hair matrix, within our routine analyses control for patients with methadone treatment or from autopsy cases. The development of the method involved an extraction technique based on incubation in concentrated sodium hydroxide (NaOH) solution, providing a dissolution of the keratin matrix.[12],[13] Hair sampling is easy to perform, not invasive, and relatively stable, moreover less affected by adulterants.[18] Hair samples allow a retrospective determination of the drug use history depending basically on hair length due to their accumulation in keratin, taking into account that head hair grows at an average rate of 1 cm circa each month,[19] and being able to confirm long-term exposure: it is, therefore, a reliable and valuable tool to assess chronic use of drugs in a specific population.[18] Moreover, parent drugs prevalently accumulate in hair and keratinized matrices in general with respect to unmetabolized drugs,[20] avoiding hydrolysis steps for the determination of metabolites.
| Subjects and Methods | |
Reagents and standards
Water, sodium dodecyl sulfate (SDS), acetone, acetonitrile, formic acid, phosphate buffer and methanol, chloroform and isopropanol, NaOH, hexane, and ethyl acetate were purchased from Sigma Aldrich, Milano, Italy. 2-Ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine (EDDP), methadone (1,1-diphenyl-1-(2-dimethylaminopropyl)-2-butanone), cannabinol (6, 6, 9-trimethyl-3-pentyl-6H-dibenzo[b, d]pyran-1-ol), cannabidiol (2-[(1R,6R)-3-methyl-6-(prop-1-en-2-yl)cyclohex-2-en-1-yl]-5-pentylbenzene-1,3-diol), Δ9-tetrahydrocannabinol (THC), UR144 (1-pentyl-1H-indol-3-yl)(2, 2, 3, 3-tetramethylcyclopropyl)methanon), CP47497 (2-[(1R,3S)-3-hydroxycyclohexyl]-5-(2-methyl-2-octanyl)phenol) and its homolog CP47497-C7, 1-([5-fluoropentyl]-1H-indol-3-yl)-(naphthalen-1-yl)methanone (AM2201), (1-hexyl-1H-indol-3-yl)-1-naphthalenyl-methanone (JWH-019), (4-methoxy-1-naphthalenyl) (1-pentyl-1H-indol-3-yl)methanone (JWH-081), (4-methyl-1-naphthalenyl) (1-pentyl-1H-indol-3-yl)methanone (JWH-122), 1-(1-pentyl-1H-indol-3-yl)-2-(2-methoxyphenyl)-ethanone (JWH-250), tetrahydrocannabinol-D3 (THC-D3), EDDP-D3, and methadone-D3 were supplied from LGC standards (Milan, Italy). Standard compounds were stored according to supplier recommendations until their use.
Calibration and sample preparation
Hair strands were collected either from routine analyses or from autopsies, cut from the posterior vertex region of the head, close to the scalp since this region is associated with least variation in growth rates (the amount required by SOHT guidelines is a pencil thickness.[18]) Hair sample aliquots were washed with 3 mL × 3 of a solution of SDS 1%, rinsed twice with 3 mL of distilled water, and then twice with 1 mL of acetone. After drying, each sample was segmented in samples of 1 cm each circa, then each one shred and grinded into small pieces of 1 mm circa (30 mg in weight).
A working solution mixture of deuterated drugs of abuse (mix drugs-Deut) containing EDDP D3, methadone D3, and THC D3 at 10 μg/mL was prepared by proper dilution of the standard solutions and stored at −20°C until use.
Individual methanolic stock solutions were used to prepare a working solution at a concentration of 10 μg/mL. Calibration curves were prepared by addition of the appropriate amount of cannabinoids to 30 mg of blank hair sample (collected from three different drug-free subjects) to obtain the following concentrations: 0.1, 0.5, 1, 2, 5, 10, and 20 ng/mg.
Thirty milligrams of hair samples was put in a vial and 3 μL mix drugs-Deut plus 500 μL of NaOH were added to digest hair sample at 90°C for 30 min. Then, the sample was extracted with 1.5 mL of a mixture composed by hexane: ethyl acetate (9:1) by automated shaking for 10 min and centrifuged for 5 min at 2000 rpm. The supernatant was then transferred into another vial and evaporated under a gentle stream of nitrogen to dryness. After the evaporation step, the sample was reconstituted with 100 μL of MeOH and 2 μL was injected in the GC/MS equipment.
Gas chromatography/mass spectroscopy equipment
The GC instrument was an Agilent 7820A (Agilent Technologies, Santa Clara, CA, USA), and the parameters chosen for optimization were the following: the liner temperature was held at 270°C; helium was used as a carrier gas at a constant flow of 40 mL/min. The oven temperature started from 100°C, then by 20°C/min was held at 250°C for 10 min, then at 20°C/min to 280°C and was held for 5 min and finally to 320°C for 4 min.
The detection system was an Agilent 5977B single quadrupole MS operating in selective ion monitoring (SIM) mode. The column was a J and W DB-5 (5% phenylmethyl silicone) capillary column (30 mm × 0.25 mm. i.d., 0.25 μm film thickness, Agilent Technologies). Characteristic ion fragments of investigated compounds were chosen and optimized injecting the individual methanolic solutions in scan mode before developing the analytical method in SIM [as reported in [Table 1]a. After pretreatment of the sample, 2 μL was injected into the instrument.
 | Click here to view |
Method validation
The method was validated according to the Food and Drug Administration guidelines[21] and was evaluated for linearity, limit of detection (LOD), limit of quantification (LOQ), lowest limit of quantitation (LLOQ), accuracy, and repeatability. The linearity of the assay was calculated by the method of least squares and expressed as coefficient of determination (R2). Calibration curves were prepared in triplicate in 3 different days by adding to blank hair samples a mixture of the commercially available standards at a concentration of 10 μg/mL in the proper amount to obtain the range of concentration and to determine the LOD and of quantification (LOQ). These parameters were studied using serial dilutions of the substances of interest in matrix in triplicate and analyzed in 5 different days. Repeatability and accuracy were assessed at three concentrations: low (quality control [QC] 1), medium (QC2), and high (QC3) injected in quintuplicate in 3 different days and were expressed, respectively, as CV% and E%. The parameters studied are listed below.
Specificity
Ten negative hair samples from voluntary subjects were collected and analyzed to determine specificity and verify, therefore, the absence of interfering peaks that could hinder the analytes. Specificity was also assessed by analyzing samples spiked with a sample at a concentration of 500 pg/mg of compounds with most common illicit or therapeutic drugs (such as cocaine and metabolites, opiates, benzodiazepines, and various antipsychotic drugs). Hence, satisfactory specificity was established if no interfering signals were found in terms of characteristic fragments and RT related to endogenous or exogenous compounds.
Limit of detection and limit of quantification
The LOD was calculated at a concentration value giving an S/N >3 for at least three ion fragments for each substance while the LOQ was considered the concentration value giving an S/N >10 for three ion fragments and acceptable accuracy and precision (%CV and %E <20%). LLOQ was calculated at the concentration value giving an S/N ratio >5. These parameters were studied using scalar dilutions of the substances of interest in hair in quintupled.
Linearity
The linearity of the method for each compound was studied in the range from the LLOQ of each substance to 20 ng/mg, performing triplicate analyses for each level. Calibration curves were built by linear regression of the area ratio of each substance with their internal standard (IS) versus the concentration of analyte.
Accuracy and precision
QC samples were prepared at three concentration levels: low (QC1), medium (QC2), and high concentrations (QC3). Accuracy and precision were assessed by analyzing the QCs in quintuplicate in three different days and were expressed respectively as %error (%E) and standard deviation (STD).
Memory effect
Memory effect, intended as carryover of analytes from sample to sample, was evaluated: two blank samples were injected after each run of spiked samples at 50 and 100 ng/mg and analyzed after positive samples.
Identification criteria
The criteria to be fulfilled for the identification of analytes were RT, the presence of three ion fragments, and their relative ion intensities. For the identification of an analyte, RT should not vary more than ±2.5%; relative ion intensities should not vary more than ±20% for ions with relative intensities >50%, ±25% for ions with relative intensities between 10% and 50%, and ±50% for ions with relative intensities <10%, with respect to a spiked control sample.
Recoveries
Recovery experiments were performed by comparing the analytical results for extracted samples at three concentrations (low 1 ng/mg, medium 10 ng/mg, and high 20 ng/mg) with samples spiked with standards after the extraction procedure that represent 100% recovery.
| Results | |
Specificity
The proposed method demonstrated its specificity for the detection and quantification of methadone and its main metabolite EDDP, including cannabinoids and the most common NPS in hair samples, verifying the absence of peaks that could interfere with the substances of interest.
Limit of detection and limit of quantification
All the analytes investigated were detectable in the range from 0.05 ng/mg to 0.5 ng/mg. The only exception was AM2201 which could be determined at 1 ng/mg [Table 1]b.
Linearity
From calibration curves, built by linear regression of the area ratio of each substance with their IS versus the concentration of analyte, the linearity of the assay was calculated by the method of least squares and expressed as coefficient of determination (R2). The method was linear in the range from LOQ to the highest concentration assessed with quadratic regression coefficients (R2) ranging from 0.9978 to 0.9997. R2 is reported for each analyzed substance in [Table 1]b.
Accuracy, precision, and recoveries
Accuracy and precision were expressed, respectively, as %CV and STD, and the results are shown in [Table 1]b. CV% values are lower than 20% for low concentrations and lower than 15% for high concentrations; therefore, according to the guidelines, the method showed acceptable accuracy and precision values. As expected after liquid/liquid extraction, a low matrix effect was observed: recovery percentages were very high (around 100%) for almost all the monitored compounds, except for JWH 019 and JWH 122, as shown in [Table 1]b.
In summary, validation parameters such as LODs, LOQs, repeatability, accuracy, and linearity were satisfactory for its application on real specimens. LODs, LOQs, R%CV, standard deviation, and the mean concentration for the analyzed compounds are reported in [Table 1]b (nominal values of QCs 1, 2, and 3 were 1 ng/mg, 5 ng/mg, and 20 ng/mg, respectively). Accuracy and repeatability were acceptable for all the analytes at their respective LOQs. Recovery experiments varied from 58.3% to 103.0%. [Figure 1] shows an extracted ion chromatogram of the quantifier ion for all the substances investigated.
 | Figure 1: Extracted ion chromatogram of the quantifier ion for all the substances investigated at a concentration of 20 ng/mg Click here to view |
Application to real samples
Since the recent application of the protocol in our laboratory, the described method was applied on 15 authentic specimens from our cases: five showed the presence of methadone and EDDP: for example, the first analyzed hair sample's segments were from a female subject found dead in her apartment from which we were able to collect 15 cm of hair strand. Nine segments (S) of 1 cm circa were prepared and analyzed and showed the following results: the root was positive for methadone at 4.31 ng/mg, 6.42 ng/mg (S-1), 5.47 ng/mg (S-2), 2.51 ng/mg (S-3), 53.19 ng/mg (S-4), 129.57 ng/mg (S-5), 361.08 ng/mg (S-6), 86.87 ng/mg (S-7), and 149.80 ng/mg (S-8), including its metabolite EDDP at 7.98 ng/mg (root), 7.22 ng/mg (S-1), 10.23 ng/mg (S-2), 7.65 ng/mg (S-3), 38.93 ng/mg (S-4), 72.38 ng/mg (S-5), 5 ng/mg (S-6) 23.47 ng/mg (S-7), and 118.47 ng/mg, respectively (S-8). In [Figure 2], chromatogram of authentic postmortem hair sample (S-5) positive for methadone and EDDP is shown at 129.57 ng/mg and 72.38 ng/mg, respectively. The subject had a well-known history of substance abuse and resulted positive to many xenobiotics in biological fluids as well. Results obtained demonstrated the proficiency of the developed method to determine, with a satisfactory sensitivity and sensibility, the drugs of abuse in hair samples involved in the study rapidly and with a simple sample pretreatment.
 | Figure 2: Chromatograms of an authentic postmortem hair sample positive for methadone and 2-ethylidene-1,5-dimethyl-3,3-diphenylpyrrolidine at 129.57 ng/mg and 72.38 ng/mg, respectively Click here to view |
| Discussion | |
Although a few immunochemical rapid tests can detect few NPS, the gold standard for their detection is chromatography coupled to mass spectrometry. Drugs levels in hair are considerably lower than those found in matrices such as blood or urine; therefore, single or tandem mass spectrometry is employed for confirmation tests. Hair testing analysis provides a retrospective timeframe via segmental analysis, due to the larger detection window when compared with other specimens (up to months, depending on strands lengths); another advantage of hair sampling is low potential for donor manipulation. In the present work, we developed and validated an analytical method to employ either after autopsy or in a population of routine analysis for methadone treatment, showing a particularly high potential for the identification of NPS users. As our study confirms, the method is suitable for analyses of studied compounds; however, there are some limitations hereby discussed. The positivity rate obtained from the study is indeed influenced by the number of samples that we could be able to collect and analyze, although it has to be stressed that these are preliminary results and that more samplings and analyses will be performed; therefore, a significant greater number of authentic cases will be assayed.
In addition, since NPS comprehends a very wide range of substances with different chemical structures, it must be noticed that the application of our analytical procedure might not be suitable for other compounds (i.e., cathinones). In particular, heating the samples as a pretreatment procedure might affect and potentially have a destructive effect on the chemical stability of NPS with low boiling points or different structures, such as mephedrone or other synthetic derivatives.[22] For example, ester analogs (e.g., PB-22) decompose (or participates in the ester-exchange reactions) in the injection port; another example is that cyclopropyl urinary metabolites (e.g., UR-144) undergo a thermal degradation mainly in GC column.[23],[24],[25] To avoid such issues, when performing splitless injection, an injector temperature of 270°C and a surface deactivated injector liner without glass wool minimizes the degradation and enhances the sensitivity. These results indicate that special attention is required for GC-MS analysis of NPS.[23] Similar mass spectra are sometimes obtained by GC–MS analyses due to regio- and ring-substituted analogs available on the market: the misidentification of these analogs arises when comparing data only with the available mass spectra. When tandem and high-resolution MS are used to identify the conformational isomers or regioisomers, such misidentification does not occur. Moreover, compounds that are thermally unstable might decompose in the GC injection port, especially those with polar groups (i.e., amino or hydroxyl groups) that can cause a polar interaction with the column stationary phase, leading to poor detection of the analyte. To overcome these problems, derivatization step should be added onto the extraction method for these compounds; however, this step would be time-consuming and potentially toxic. From our validation study, satisfactory results on our set of analytes can be obtained overcoming the derivatization step.
Indeed, synthetic cannabinoids are not ideal compounds for GC-MS analysis because they are neutral to weakly acidic compounds and have high molecular weight. However, concerning the substances taken into consideration in the validation of the presented method, no destructive effect was noticed; therefore, the method is applicable only for the relevant compounds included in this study and more experiments will be performed when other NPS will be included in the procedure. The phenomenon might interest some metabolites that, however, are not the focus of our study, because the parent compounds are the main targets in hair analyses.
Hair analysis for NPS is still at an early stage of development, particularly on the toxicological screening side. The proposed method allows for the identification of synthetic cannabinoids, cannabinoids, and methadone (including its main metabolite) in hair samples with a simple sample pretreatment.
| Conclusions | |
The identification of NPS in biological samples is one of the emerging challenges for forensic laboratories due to the necessity of detection and confirmation of a very large class of substances, often not structurally correlated. Due to the reoccurring threat of synthetic cannabinoids to public health and their rapidly increasing abuse worldwide, it is necessary to develop reliable analytical methods for their detection in different biological matrices. SCs are constantly being modified and rapidly becoming widely available; therefore, laboratories should update their scope for detecting the most prevalent compounds at specific times.
Blood and urine are the first choice of sample for testing; however, hair is often used as an alternative matrix in repeated drug exposure. The method validation presented herein is a straightforward, selective, and accurate method for the determination of some drugs belonging to the CP and aminoalkylindole structural classes. The described method can be useful not only in the forensic investigation of NPS-related addiction histories but also in epidemiological and retrospective studies on the spread of NPS among specific safety-sensitive social workers, such as drivers.
This study has been approved by the local ethics committee and the consent was exempted.
