Journal of Analytical Toxicology

Vignali,, Claudia;Moretti,, Matteo;Groppi,, Angelo;Osculati, Antonio Marco, Maria;Tajana,, Luca;Morini,, Luca

doi: 10.1093/jat/bky047pmid: 30060126

Abstract We report the analysis of the synthetic cathinones α-pyrrolidinohexiophenone (α-PHP) and α-pyrrolidinopentiophenone (α-PVP), both pyrovalerone derivatives, in blood, urine, gastric contents, main tissues and hair of a deceased person. Qualitative and quantitative analyses were performed by LC–MS-MS. All the biological samples were collected during autopsy and extracted/purified onto a solid phase extraction cartridge before instrumental analysis. The method was validated for blood and urine and proved to be highly sensitive and specific for both the synthetic cathinones (limit of detection: 0.2 ng/mL and limit of quantification: 0.5 ng/mL). Analyses provided negative results for α-PVP in all biological samples except for the 2-cm proximal hair segment, confirming that the young man had consumed in the last 2 months this compound; instead hair analysis proved that the man was a heavy α-PHP user. α-PHP was identified and quantified in biological fluids and tissues. Interestingly, bile and urine concentrations (1.2 and 5.6 ng/mL, respectively) were fairly lower than blood collected into the thoracic cavity (15.3 ng/mL). The highest concentrations were measured for lung (71.1 ng/mL) and spleen (83.8 ng/mL). Concentrations of 3.5, 7.9, 4.7 and 23.6 ng/mL were measured in liver, kidney, brain and heart, respectively. Even if it is not possible to evaluate the presence of this drug in biological samples as a cause of death, to our knowledge, this is the first case of α-PHP finding in postmortem samples, and its potential toxic effects should be elucidated in the future. Introduction Synthetic cathinones, also called “bath salts”, are new psychoactive substances (NPS) that in recent years have become popular drugs of abuse. These are phenylalkylamine derivatives and the synthesis of these compounds has been reported since 1920s; initially used for medical purposes. Years later, at the end of 2000s, the interest in synthetic cathinones as psychoactive drugs grew up due to their large availability on the Internet. The primary reason for so large spread in Europe was that these drugs were initially legal. After the introduction of regulatory measures restricting the sale of synthetic cathinones, consumers were forced to stop buying them on the Internet and began purchasing from local dealers. Synthetic cathinones were the most frequently seized NPS in Europe in 2015, with over 25,000 seizures. Use of synthetic cathinones has been reported in 15 European countries, with a large variability of compounds by country; few people currently enter treatment in Europe for problems associated with use of these drugs, although under-reporting is likely (1). Synthetic cathinones are the most frequently seized NPS in Italy; among them, 3-methylmethcathinone, 4-methylethcathinone and methylenedioxypyrovalerone (MDPV) were often identified, but also α-PHP and α-PVP findings were reported (2). Synthetic cathinones can be ingested, snorted, smoked or, more rarely, injected. Desired effects of these compounds include empathy, increased energy and libido; negative effects, mainly cardiac and psychiatric, may occur. Some cases of intoxication associated with use of synthetic cathinones have been reported (3–6). Synthetic cathinones have variable effects and potency levels on serotonin, dopamine and noradrenaline pathways, but typically possess sympathomimetic/amphetamine-like effects (7, 8). Instead, pyrovalerone derivatives are reported to show a cocaine-like mechanism of action (as dopamine re-uptake inhibiting agents): α-PHP is described to be more potent than MDPV and α-PVP, that show similar potency (9). α-PHP was identified for the first time in 2014 in seized materials in Japan (10). The molecule is chemically related to α-PVP, a compound that had been used as appetite suppressant and in the treatment of chronic fatigue, having an extra carbon on the alkyl side chain (Figure 1). There are a few published data on pharmacokinetics and pharmacodynamics of α-PHP, but in a recently published article, the structures of some metabolites were tentatively elucidated, analyzing urine samples of a drug abuser. α-PHP is extensively metabolized (19 phase I metabolites and 9 glucuronide conjugates were identified). Nevertheless, the parent drug is detectable in urine sample in large amount (11). Concentrations of α-PHP in urine were measured in the range 1–300,000 ng/mL and in blood in the range 4–10 ng/mL (12). Plasmatic half-life of α-PHP was investigated by Fuyita et al.: they analyzed 14 blood samples, collected subsequently from an intoxicated patient. Due to the long-term clinical symptoms of the patient induced by α-PHP, a long half-life was suspected, and in fact it was found to be ~37 h (13). In recent years, few cases of pyrovalerone derivatives-related intoxications and deaths have been reported (14–16). In these cases, blood concentrations were extremely variable, but it should be considered that exposure to the drug may be highly erratic, because users cannot know the real content of doses and their purity. Case Report A 27-year-old man was admitted to a local psychiatric hospital for drug abuse evaluation, clinical stabilization and withdrawal treatment, after having suffered from psychomotor agitation. He had a long history of polysubstance abuse and addiction. Upon arrival at the hospital, the patient was conscious, oriented and cooperative. He showed little postural instability and some difficulties with fine motor skills. Cardiovascular and respiratory parameters were normal. A urine sample was collected and screened for common drugs of abuse (opiates, cannabinoids, cocaine, amphetamines, benzodiazepines). The test provided positive results for benzodiazepines only. According to self-declaration, the young man started using cannabis when he was 14 and continued until the age of 21. During the same period, he also smoked and snorted cocaine, at first rarely and then daily. He stopped using cocaine for 1 year, but simultaneously began to binge-drink alcohol. He also said that he often used ketamine around the age of 20. Occasionally he took LSD, but with this drug he experienced terrifying hallucinations. After these episodes, he started suffering from psychiatric disorders, requiring multiple admissions to psychiatric hospitals and being treated with antipsychotics and antidepressants. He then used amphetamines consistently for several years, too. In order to treat psychiatric disturbances, during the first day of hospitalization, the young man was treated with diazepam, clotiapine e quetiapine. His condition, however, did not improve: he reported that he was not able to leave his room because just the motion of his head gave him hallucinations. The following day, soon after suffering withdrawal symptoms, the patient admitted that in the last few months he frequently smoked crystal-like drugs, purchased online and referred to as “cathinones”. After taking those substances, he suffered from adverse effects (such as bone and muscular pains, agitation, aggression and visual hallucinations). He also developed a strong and irrepressible compulsion to redose, as well as addiction/dependence. He tried without success to reduce these symptoms firstly by taking benzodiazepines (without any medical supervision), and then by smoking heroin and using illegally obtained methadone. Based on the clinical records provided by the patient, opioid withdrawal was suspected and physicians decided to submit the young man to a methadone treatment. The starting dose was 10 mg, which was increased to 50 mg within 2 days. After 3 days, a nurse reported that the patient showed signs of sedation. A few hours later, the man was found dead in his bed. His body was subjected to a full autopsy by forensic pathologists. Autopsy revealed massive cerebral and pulmonary edema, visceral congestion and left ventricular hypertrophy. Materials and Methods Chemicals α-PHP, α-PHP, mephedrone-D3 and methadone were obtained by LGC Standards (Milan, Italy). Quetiapine, clotiapine, diazepam and desmethyldiazepam standards were purchased from Sigma-Aldrich (Milan, Italy). Formic acid for mass-spectrometry was obtained from Sigma-Aldrich (Milan, Italy). HPLC-grade methanol, acetonitrile, dichloromethane, isopropanol and ammonia were purchased from Mallinkrodt Baker (Milan, Italy). Bond Elut Certify I cartridges were purchased from Agilent (Milan, Italy). Mobile phase components were 0.1% formic acid (A) and methanol (B). Sample treatment Working solutions of α-PHP and α-PVP were prepared using methanol at a concentration of 1 μg/mL for mass-spectrometry tuning and selectivity experiments. Calibrators were carried out by mixing the stock solutions and diluting with methanol at the following concentrations: 0.001, 0.005, 0.010, 0.050 and 0.100 μg/mL. All solutions were stored at −20°C. The calibration curves were prepared using pooled blank postmortem blood and urine samples. All samples used for calibration were checked before analysis. Tissues were homogenized using a Precellys Evolution, Bertin (AlfaTech, GE, Italy) before sample treatment. About 500 μL body fluids (blood, urine, bile and gastric contents) and 0.5 g homogenized tissue samples (liver, kidney, brain, lung, spleen and heart) were mixed with 2 mL phosphate buffer at pH 6.0 and 0.02 mL of mephedrone-D3 (internal standard—IS), at the concentration of 0.001 μg/mL. A hair sample (2-cm proximal) was washed with 1 mL methylene chloride and 1 mL methanol, pulverized using a Precellys Evolution and weighed (50 mg). Then, 0.02 mL of mephedrone-D3 at the concentration of 0.001 μg/mL were added, together with 1 mL HCl 0.1 N and overnight incubation at 37°C was performed. After centrifugation, samples were purified on solid phase extraction (SPE) cartridges (Bond Elut Certify, 130 mg, Agilent Technologies, MI, Italy). The columns were initially conditioned with 2 mL methanol and equilibrated with 2 mL phosphate buffer at pH 6.0; after loading the samples, the cartridges were washed with 2 mL water, 3 mL HCl 0.1 M and 5 mL methanol. Finally, synthetic cathinones were eluted from the cartridges with 2 mL dichloromethane: isopropanol (8:2) solution containing 2% ammonia. The eluates were taken to dryness under nitrogen stream at 35°C and the samples were reconstituted in 150 μL mobile phase. About 10 μL were injected in the LC–MS-MS system. Urine and blood samples were previously analyzed as blank samples. LC–MS-MS settings The method was developed with an Agilent 1100 Series system (vacuum degasser, binary pump and column compartment) and an Agilent 1200 Series isocratic pump and autosampler maintained at 4°C (Agilent Technologies, Palo Alto, CA, USA) coupled with a 4000 QTrap mass spectrometer (AB/ Sciex, Foster City, CA, USA). The LC injector needle was externally washed with methanol prior to any injection. A kinetex C18 column (100 × 2.1 mm i.d., 5 μm particle size, Phenomenex, Castelmaggiore, Italy) was kept at 25°C during the chromatographic run. Flow rate (0.3 mL/min) was set at gradient mode as follows: from 95% A to 5% in 6 min, maintained at 5% for 3 min and reequilibrated for 8 min. The ESI source settings were: source temperature, 500°C; nebulization and heating gas (air): 30 psig and 25 psig, respectively; curtain gas (nitrogen): 20 psig. Ion-spray voltage was set at 5,000 V in positive ion mode. Mass analysis was performed in Multiple Reaction Monitoring (MRM) mode and positive polarization. All parameters, including declustering potential (DP), collision energy (CE) and cell exit potential (CXP) were separately optimized for each analyte by infusing in the mass spectrometer a 1-μg/mL solution in 0.1% formic acid (Table I). Two transitions for each analyte were chosen for identification while the most intense was selected for quantification purposes. Table I. MRM transitions and parameters Analyte Q1 (m/z) Q3 (m/z) DP (V) EP (V) CE (eV) α-PHP 246.4 105.2 80 10 40 α-PHP 246.4 140.2 80 10 35 α-PVP 232.1 91.3 90 10 65 α-PVP 232.1 105.3 90 10 31 Mephedrone-D3 181.2 148.1 60 10 29 Mephedrone-D3 181.2 163.3 60 10 29 Analyte Q1 (m/z) Q3 (m/z) DP (V) EP (V) CE (eV) α-PHP 246.4 105.2 80 10 40 α-PHP 246.4 140.2 80 10 35 α-PVP 232.1 91.3 90 10 65 α-PVP 232.1 105.3 90 10 31 Mephedrone-D3 181.2 148.1 60 10 29 Mephedrone-D3 181.2 163.3 60 10 29 Table I. MRM transitions and parameters Analyte Q1 (m/z) Q3 (m/z) DP (V) EP (V) CE (eV) α-PHP 246.4 105.2 80 10 40 α-PHP 246.4 140.2 80 10 35 α-PVP 232.1 91.3 90 10 65 α-PVP 232.1 105.3 90 10 31 Mephedrone-D3 181.2 148.1 60 10 29 Mephedrone-D3 181.2 163.3 60 10 29 Analyte Q1 (m/z) Q3 (m/z) DP (V) EP (V) CE (eV) α-PHP 246.4 105.2 80 10 40 α-PHP 246.4 140.2 80 10 35 α-PVP 232.1 91.3 90 10 65 α-PVP 232.1 105.3 90 10 31 Mephedrone-D3 181.2 148.1 60 10 29 Mephedrone-D3 181.2 163.3 60 10 29 Validation The analytical procedures have been validated according to the guidelines by Peters et al. (17). The following parameters were considered for method validation: linearity, sensitivity, specificity, accuracy, imprecision, recovery, carryover, matrix effects, and freeze and thaw stability. Linearity was evaluated using a six-point calibration curve in the range 0.5–200.0 ng/mL (calibrators: 0.5, 1.0, 10.0, 50.0, 100.0, 200.0 ng/mL). Sensitivity, expressed as limit of detection (LOD) and lower limit of quantification (LLOQ), was evaluated for each analyte using decreasing concentrations of analyte in drug-fortified blood and urine. LOD was defined as the lowest concentration with neat peaks and a signal-to-noise ratio of at least 3, and a relative retention time (RRT) within ±2% of the average calibrator (10 ng/mL) RRT. LLOQ was defined as the lowest concentration with the same parameters of LOD and with signal-to-noise ratio of at least 10 and acceptable accuracy and precision as defined below. Accuracy and imprecision were measured at two quality control (QC) levels (5.0 and 200 ng/mL). The lower QC level was chosen according to the results obtained on authentic samples. Spiked blank tissues at QC levels were evaluated on calibration curve prepared with blood. Accuracy was calculated as the percentage deviation of the mean calculated concentration (n = 10 over a 4-day period; absolute value) from the corresponding nominal value. Intra-assay and inter-assay imprecisions were calculated as the coefficient of variation (CV%) of five replicates for each QC level. Specificity was evaluated by adding high concentrations (1,000 ng/mL) of potentially interfering licit and illicit drugs to calibration points and QC samples. Traditional drugs of abuse, benzodiazepines, cardiovascular drugs, antidepressants, antipsychotics and metabolites were included in the drug-fortified samples. Ten blank samples of urine, blood and liver were also processed to exclude interferences from endogenous substances. Interference was excluded if samples were within ±20% target concentration. Absolute analytical recovery was assessed for each analyte using three replicates for each QC sample concentration by comparing the peak areas obtained when samples were analyzed by adding the QC samples and the ISs in the extract of blank blood and urine prior to and after the extraction procedure. Carryover was assessed by injecting extracted blank blood and urine samples immediately after analysis of the highest concentration point of the calibration curve on each of the days of the validation protocol and measuring the area of eventual peaks, present at the RRTs of analytes under investigation. Matrix effects were measured by comparing blank biological samples with water samples, spiked at QC levels and processed with the described procedure; absolute peak area in blood, urine and in water were compared. Experiments were carried out in triplicate, using 10 different blank samples. Results were calculated by comparing the peak areas of α-PHP and α-PVP in biological samples versus those measured in water. The effect of three freeze-thaw cycles (storage at −20°C) on the compounds stability in blood and urine was evaluated by repeated analysis (n = 3) of QC samples. In consideration of the fact that the procedure for detecting α-PHP and α-PVP in hair was developed for screening purposes, validation was limited to the verification of method selectivity (seven different blanks injected), to the determination of the LODs. Further toxicological analyses Quantitative determination of ethanol, and systematic toxicological analysis (STA) to detect acidic, neutral and basic drugs including those that are commonly prescribed, non-prescribed and illicit drugs were performed on blood sample: STA was carried out using gas chromatography-mass spectrometry (GC–MS) technique preceded by a SPE with mixed-mode cartridges Bond Elut Certify, using the method detailed in a previously published study (18). GC–MS data files were processed by an original procedure for the automated purification of mass spectra from the total ion chromatogram (19). Quantitative analysis of methadone in blood was performed by GC–MS using the method routinely used in our laboratory. Quantitative analysis of diazepam, desmethyldiazepam, clotiapine and quetiapine in blood were carried out by LC–MS-MS, using the method detailed in a previously published study (20). Results GC–MS and LC–MS-MS analyses revealed the presence of methadone (440.0 ng/mL), diazepam (500.0 ng/mL), desmethyldiazepam (340.0 ng/mL), clotiapine (360.0 ng/mL) and quetiapine (180.0 ng/mL) in blood sample. Ethanol intake was excluded. Linear calibration curves showed determination coefficients (R2) higher than 0.99 for both α-PHP and α-PVP. LODs were set at 0.2 ng/mL for both α-PHP and α-PVP in urine and blood while the lowest calibration point (0.5 ng/mL) was chosen as LLOQs for the two cathinones in biological samples (Table II). A LOD of 10 pg/mg was measured in hair for both α-PHP and α-PVP. Table II. Validation parameters Analyte QC concentration (ng/mL) Accuracy (RSD%) Intraday-imprecision (CV%) Interday-imprecision (CV%) Recovery (%) Matrix effects (%) Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine α-PHP 5.0 5.0 13.4 2.4 12.1 10.4 15.4 16.2 97.8 94.5 −17.8 −18.5 200.0 200.0 4.8 3.5 9.7 10.2 10.3 9.5 91.0 95.4 −12.1 −10.1 α-PVP 5.0 5.0 15.7 17.9 11.5 15.3 16.3 18.4 87.4 89.6 −16.6 −18.4 200.0 200.0 8.7 6.7 10.3 9.6 8.7 6.5 83.2 84.7 −14.6 −16.7 Analyte QC concentration (ng/mL) Accuracy (RSD%) Intraday-imprecision (CV%) Interday-imprecision (CV%) Recovery (%) Matrix effects (%) Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine α-PHP 5.0 5.0 13.4 2.4 12.1 10.4 15.4 16.2 97.8 94.5 −17.8 −18.5 200.0 200.0 4.8 3.5 9.7 10.2 10.3 9.5 91.0 95.4 −12.1 −10.1 α-PVP 5.0 5.0 15.7 17.9 11.5 15.3 16.3 18.4 87.4 89.6 −16.6 −18.4 200.0 200.0 8.7 6.7 10.3 9.6 8.7 6.5 83.2 84.7 −14.6 −16.7 Table II. Validation parameters Analyte QC concentration (ng/mL) Accuracy (RSD%) Intraday-imprecision (CV%) Interday-imprecision (CV%) Recovery (%) Matrix effects (%) Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine α-PHP 5.0 5.0 13.4 2.4 12.1 10.4 15.4 16.2 97.8 94.5 −17.8 −18.5 200.0 200.0 4.8 3.5 9.7 10.2 10.3 9.5 91.0 95.4 −12.1 −10.1 α-PVP 5.0 5.0 15.7 17.9 11.5 15.3 16.3 18.4 87.4 89.6 −16.6 −18.4 200.0 200.0 8.7 6.7 10.3 9.6 8.7 6.5 83.2 84.7 −14.6 −16.7 Analyte QC concentration (ng/mL) Accuracy (RSD%) Intraday-imprecision (CV%) Interday-imprecision (CV%) Recovery (%) Matrix effects (%) Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine Blood Urine α-PHP 5.0 5.0 13.4 2.4 12.1 10.4 15.4 16.2 97.8 94.5 −17.8 −18.5 200.0 200.0 4.8 3.5 9.7 10.2 10.3 9.5 91.0 95.4 −12.1 −10.1 α-PVP 5.0 5.0 15.7 17.9 11.5 15.3 16.3 18.4 87.4 89.6 −16.6 −18.4 200.0 200.0 8.7 6.7 10.3 9.6 8.7 6.5 83.2 84.7 −14.6 −16.7 Accuracy, intra- and inter-assay imprecisions were within the established acceptance criteria (Table II). CV% and BIAS% were lower than 15% for QC levels prepared using blank tissues. No additional peaks due to endogenous substances that could have interfered with the detection of the two synthetic cathinones have been observed. Likewise, none of the drugs of abuse or abovementioned medicines, carried through the entire procedure, interfered with the assay and with the accurate quantification of the low QC samples. Analytical recoveries obtained after extraction procedure for the two different QC samples ranged between 83.2% and 97.8% for both analytes (Table II). No carryover was observed when blank samples were injected after the highest point of the calibration curve. Matrix effects were negligible. No relevant degradation was observed after any of the three freeze/thaw cycles, with differences in the initial concentration <10%. All the biological samples collected during autopsy provided negative results for α-PVP except for the 2-cm proximal hair segment, confirming that the young man had consumed in the last 2 months this synthetic cathinone. α-PHP was identified and quantified in all the biological fluids and tissues. Results are listed in Table III. The absolute peak area of α-PHP measured in hair sample was extremely high, suggesting that the subject was probably a heavy consumer of α-PHP. However, it cannot be excluded a potential contamination due to sweat (21). Table III. Postmortem samples concentrations Biological sample α-PHP ng/mL(g) Blood 15.3 Urine 5.6 Bile 1.2 Gastric contents pos Liver 3.5 Kidney 7.9 Spleen 83.8 Lung 71.1 Brain 4.7 Heart 23.6 Hair 1078.0a Biological sample α-PHP ng/mL(g) Blood 15.3 Urine 5.6 Bile 1.2 Gastric contents pos Liver 3.5 Kidney 7.9 Spleen 83.8 Lung 71.1 Brain 4.7 Heart 23.6 Hair 1078.0a apg/mg. Table III. Postmortem samples concentrations Biological sample α-PHP ng/mL(g) Blood 15.3 Urine 5.6 Bile 1.2 Gastric contents pos Liver 3.5 Kidney 7.9 Spleen 83.8 Lung 71.1 Brain 4.7 Heart 23.6 Hair 1078.0a Biological sample α-PHP ng/mL(g) Blood 15.3 Urine 5.6 Bile 1.2 Gastric contents pos Liver 3.5 Kidney 7.9 Spleen 83.8 Lung 71.1 Brain 4.7 Heart 23.6 Hair 1078.0a apg/mg. Discussion Analysis of NPS in biological samples may be difficult. Few laboratories own the instrumentation required for their identification and an adequate number of analytical standards, given the large number of molecules belonging to this category; any finding of drug specimens may indeed be of great help. In our case, suspicion that the man could have taken exactly α-PHP, came from the discovery of a plastic wrap containing some small crystals, identified as this compound, in the apartment where he lived. The case here reported involve a young man that started using synthetic cathinones after previous experiences with cannabis, LSD and cocaine, and in a short time he could not stop taking them. After a few months of intake, he was in a recurrent state of agitation with visual hallucinations and terrifying nightmares, so his parents convinced him to go into rehab. It is recognized that synthetic cathinones have both stimulant and psychoactive effects: self-reported symptoms associated with their use include cardiovascular effects (palpitation, shortness of breath), gastrointestinal (nausea, vomiting, abdominal pain), neurologic (aggressiveness, dizziness, headache, memory loss, tremor, seizures), ophthalmologic (blurred vision, mydriasis), psychological (anger, anxiety, hallucinations, depression, dysphoria, panic, nightmares, paranoia) (22). A significant proportion of synthetic cathinones users report tolerance, dependence or withdrawal symptoms (23); many of them experienced strong craving to repeat doses after taking the drug. Long-term effects related to the intake of these substances are typically associated with the imbalance of a range of neurotransmitter pathways/receptors, and consequently with the risk of psychopathological disturbances (24). In order to treat psychiatric disturbances, the man was given diazepam, quetiapine and clotiapine, but his condition did not improve. All these drugs were detected in blood sample at therapeutic levels. Furthermore, methadone was administered because the patient said he used to take heroin and methadone to reduce the side effects of cathinones, but unexpectedly, morphine and methadone were not detected in hair sample collected during autopsy. Methadone blood concentration (440.0 ng/ml) could justify the death in a naïve subject, according to current scientific knowledge, so methadone was determined to be the cause of death. Other drugs (diazepam, clotiapine and quetiapine) due to the depressing effect on the nervous system, may had contributed to the man’s death. The finding of α-PHP in biological fluids and tissues of the patient three days after the hospitalization is an interesting detail, and two different scenarios may be considered in trying to interpret analytical results. A concentration of 1,078.0 pg/mg α-PHP was measured in the 2-cm proximal hair segment. Though, to the best of our knowledge, this is the very first case where α-PHP was measured in hair, similar concentrations of synthetic cathinones were already measured in other cases of chronic abusers of cathinone and amphetamine derivatives (25). Hence, the level of α-PHP is more consistent with a continuing heavy consumption of the drug by the subject during, at least, the last two months before death, rather than a contamination through sweat. The man experienced compulsive redosing, so it is possible that he may have brought some doses with him. Yet, qualitative analysis of gastric contents showed the presence of α-PHP, thus suggesting a very recent consumption. Likewise, the highest concentrations of the drug were found in lung and spleen samples, both affected by massive congestion, with a higher amount of blood accumulated in these tissues. Unfortunately, quantitative analysis performed on blood sample has to be considered scarcely reliable: in fact, the analysis was carried out on blood collected into the thoracic cavity, due to the lack of peripheral and cardiac blood. According to this hypothesis, the man could have taken the drug shortly before death, perhaps to counteract the narcotic properties of methadone. Figure 1. View largeDownload slide Chemical structure of α-pyrrolidinohexiophenone (A) and α-pyrrolidinopentiophenone (B). Figure 1. View largeDownload slide Chemical structure of α-pyrrolidinohexiophenone (A) and α-pyrrolidinopentiophenone (B). The second option is that the young man could have taken the last dose before hospitalization: indeed α-PHP is thought to be eliminated from the body over a period of 150 h. A recent study reported a blood concentration of α-PHP of 175.0 ng/mL in an intoxicated patient that slowly decreased to 15.7 ng/mL after 96 h (13). Authors assume that the long elimination of α-PHP may be attributed to the gradual release of the compound from other body tissues. This hypothesis could also be consistent with our results: in this instance, the finding of α-PHP in gastric contents could be interpreted as gastric secretion of the drug, and postmortem redistribution cannot be excluded. Conclusion Synthetic cathinones have been the most frequently seized NPS in Europe in the last few years, but their finding in biological samples remains an analytical challenge, especially when use is not suspected. Furthermore, there are few studies concerning distribution of these drugs in man. To our knowledge, this is the first report about detection and quantification of α-PHP in body fluids and tissues in a postmortem case. Even if the drug was not directly the cause of death, our data may be useful for further analyses and interpretation of toxicological cases involving α-PHP. References 1 EMCDDA . ( 2017 ) European Monitoring Centre for Drugs and Drug Addiction. Trends and development. EMCDDA, Lisbon, November 2011. http://www.emcdda.europa.eu/system/files/publications/4541/TDAT17001ENN.pdf (accessed February 14, 2018) 2 Odoardi , S. , Saverio Romolo , F. , Strano Rossi , S. ( 2016 ) A snapshot on NPS in Italy: distribution of drugs in seized materials analysed in an Italian forensic laboratory in the period 2013–2015 . Forensic Science International , 265 , 116 – 120 . Google Scholar Crossref Search ADS PubMed 3 Guirguis , A. , Corkery , J.M. , Stair , J.L. , Kirton , S.B. , Zloh , M. , Schifano , F. ( 2017 ) Intended and unintended use of cathinone mixtures . Human Psychopharmacology , 32 , 1 – 17 . Google Scholar Crossref Search ADS 4 Fujita , Y. , Koeda , A. , Fujino , Y. , Onodera , M. , Kikuchi , S. , Niitsu , H. , et al. . ( 2015 ) Clinical and toxicological findings of acute intoxication with synthetic cannabinoids and cathinones . Acute Medicine & Surgery , 28 , 230 – 236 . 5 Hall , C. , Heyd , C. , Butler , C. , Yarema , M. ( 2014 ) “Bath salts” intoxication: a new recreational drug that presents with a familiar toxidrome . Canadian Journal of Emergency Medicine , 16 , 171 – 176 . Google Scholar PubMed 6 Pourmand , A. , Mazer-Amirshahi , M. , Chistov , S. , Li , A. , Park , M. ( 2018 ) Designer drugs: review and implications for emergency management . Human & Experimental Toxicology , 37 , 94 – 101 . Google Scholar Crossref Search ADS PubMed 7 Schifano , F. ( 2015 ) NPS: clinical and pharmacological issues . Drug and Alcohol Today , 15 , 21 – 27 . Google Scholar Crossref Search ADS 8 Schifano , F. Novel psychoactive substances also known as ‘legal highs’. In: Davies , S.C. (ed). Annual report of the Chief Medical Officer 2013, Public Mental Health Priorities: Investing in the Evidence . Department of Health : London , 2014 ; 259 . Chapter 16. 9 Kolanos , R. , Sakloth , F. , Jain , A.D. , Partilla , J.S. , Baumann , M.H. , Glennon , R.A. ( 2015 ) Structural modification of thedesigner stimulant α-pyrrolidinovalerophenone (α-PVP). Influences potency at dopamine transporters . ACS Chemical Neurosciences , 6 , 1726 – 1731 . Google Scholar Crossref Search ADS 10 Uchiyama , N. , Shimokawa , Y. , Kawamura , M. , Kikura-Hanajiri , R. , Hakamatsuka , T. ( 2014 ) Chemical analysis of a benzofuran derivative, 2-(2-ethylaminopropyl)benzofuran(2-EAPB), eight synthetic cannabinoids, five cathinone derivatives, and five other designer drugs newly detected in illegal products . Forensic Toxicology , 32 , 266 – 281 . Google Scholar Crossref Search ADS 11 Paul , M. , Bleicher , S. , Guber , S. , Ippisch , J. , Polettini , A. , Schultis , W. ( 2015 ) Identification of phase I and II metabolites of the new designer drug α-pyrrolidinohexiophenone (α-PHP) in human urine by liquid chromatography quadrupole time-of-flight mass spectrometry (LC-QTOF-MS) . Journal of Mass Spectrometry , 50 , 1305 – 1317 . Google Scholar Crossref Search ADS PubMed 12 Back , O. , Bäckberg , M. , Signell , P. , Helander , A. ( 2018 ) Intoxications in the STRIDA project involving a panorama of psychostimulant pyrovalerone derivatives, MDPV copycats . Clinical Toxicology , 56 , 256 – 263 . Google Scholar Crossref Search ADS PubMed 13 Fujita , Y. , Mita , T. , Usui , K. , Kamijo , Y. , Kikuchi , S. , Onodera , M. , et al. . ( 2018 ) Toxicokinetics of the synthetic cathinone α-pyrrolidinohexanophenone . Journal of Analytical Toxicology , 28 , 1 – 5 . Google Scholar Crossref Search ADS 14 Kudo , K. , Usumoto , Y. , Kikura-Hanajiri , R. , Sameshima , N. , Tsuji , A. , Ikeda , N. ( 2015 ) A fatal case of poisoning related to new cathinone designer drugs, 4-methoxy PV8, PV9, and 4-methoxy PV9, and a dissociative agent, diphenidine . Legal Medicine , 17 , 421 – 426 . Google Scholar Crossref Search ADS PubMed 15 Wyman , J.F. , Lavins , E.S. , Engelhart , D. , Armstrong , E.J. , Snell , K.D. , Boggs , P.D. , et al. . ( 2013 ) Postmortem tissue distribution of MDPV following lethal intoxication by “bath salts” . Journal of Analytical Toxicology , 37 , 182 – 185 . Google Scholar Crossref Search ADS PubMed 16 Eiden , C. , Mathieu , O. , Cathala , P. , Debruyne , D. , Baccino , E. , Petit , P. , et al. . ( 2013 ) Toxicity and death following recreational use of 2-pyrrolidino valerophenone . Clinical Toxicology , 51 , 899 – 903 . Google Scholar Crossref Search ADS PubMed 17 Peters , F.T. , Drummer , O.H. , Musshoff , F. ( 2007 ) Validation of new methods . Forensic Science International , 165 , 216 – 224 . Google Scholar Crossref Search ADS PubMed 18 Polettini , A. , Groppi , A. , Vignali , C. , Montagna , M. ( 1998 ) Fully-automated systematic toxicological analysis of drugs, poisons, and metabolites in whole blood, urine, and plasma by gas chromatography-full scan mass spectrometry . Journal of Chromatography B: Biomedical Sciences and Applications , 713 , 265 – 279 . Google Scholar Crossref Search ADS 19 Polettini , A. ( 1996 ) A simple automated procedure for the detection and identification of peaks in gas chromatography–continuous scan mass spectrometry . Application to systematic toxicological analysis of drugs in whole human blood. Journal of Analytical Toxicology , 20 , 579 – 586 . 20 Sempio , C. , Morini , L. , Vignali , C. , Groppi , A. ( 2014 ) Simple and sensitive screening and quantitative determination of psychoactive drugs and their metabolites in blood through LC-MS/MS: application on postmortem samples . Journal of Chromatography B , 970 , 1 – 7 . Google Scholar Crossref Search ADS 21 Cairns , T. , Hill , V. , Schaffer , M. , Thistle , W. ( 2004 ) Removing and identifying drug contamination in the analysis of human hair . Forensic Science International , 145 , 97 – 108 . Google Scholar Crossref Search ADS PubMed 22 Prosser , J.M. , Nelson , L.S. ( 2012 ) The toxicology of bath salts: a review of synthetic cathinones . Journal of Medical Toxicology , 8 , 33 – 42 . Google Scholar Crossref Search ADS PubMed 23 Schifano , F. , Albanese , A. , Fergus , S. , Stair , J.L. , Deluca , P. , Corazza , O. , et al. . ( 2011 ) Mephedrone (4-methylmethcathinone; ‘meow meow’): chemical, pharmacological and clinical issues . Psychopharmacology , 214 , 593 – 602 . Google Scholar Crossref Search ADS PubMed 24 Schifano , F. , Orsolini , L. , Duccio Papanti , G. , Corkery , J.M. ( 2015 ) Novel psychoactive substances of interest for psychiatry . World Psychiatry : Official Journal of the World Psychiatric Association (WPA) , 14 , 15 – 26 . Google Scholar PubMed 25 Salomone , A. , Gazzilli , G. , Di Corcia , D. , Gerace , E. , Vincenti , M. ( 2016 ) Determination of cathinones and other stimulant, psychedelic, and dissociative designer drugs in real hair samples. Analytical . Bioanalytical Chemistry , 408 , 2035 – 2042 . Google Scholar Crossref Search ADS © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Feldhammer,, Matthew;Saitman,, Alec;Nguyen,, Ly;Milstid,, Bryan

doi: 10.1093/jat/bky059pmid: 30192938

Abstract Adulteration of samples submitted for toxicological analyses can present unique challenges to non-forensic clinical laboratories. With the number of overdose-related deaths expected to surpass 60,000 in 2018, it is incumbent on all members of the healthcare team to be active participants in curbing opioid dependence and identifying prescription drug misuse and diversion. Recently published guidelines have sought to provide guidance to laboratories overseeing prescription drug-monitoring programs. We present a case of sample adulteration in an attempt to conceal prescription non-compliance. The patient possessed only an active prescription for hydrocodone but on initial antibody-based screening the sample tested positive for benzodiazepines and oxycodone in addition to opiates. Active communication between the pain management clinic and the clinical laboratory alerted staff to conduct a more thorough investigation including sample validity testing, analyses of paired serum specimens by liquid chromatography tandem mass spectrometry. Analyses revealed the patient submitted a dilute urine specimen with a crushed hydrocodone pill inside in an attempt to hide prescription non-compliance. Previous screenings had been consistent with the medication list raising the question of whether this was an isolated incident or the patient had simply been more successful in manipulating specimens in the past. This case highlights the need for good communication among all members of the healthcare team and the widespread implementation of specimen validity testing for any laboratory that receives samples from pain clinics. Introduction Widespread abuse of prescription opioids is a major contributing factor to the current epidemic in the USA. Opioid overdose has surpassed all other causes of morbidity for adults under age 50, with deaths likely to top 60,000 in 2017 alone. Approximately half of all opioid-related overdoses involve a prescription opioid, and the majority of users report the source to be friends or family with legally obtained prescriptions (1). Diversion of legally prescribed opiates, therefore, is of major concern from both a law enforcement and a public health perspective. The clinical laboratory plays an integral role as a part of the multifaceted and integrated approach to ensuring medication compliance. Patients on chronic opioid therapy are routinely enrolled in controlled substance agreements with providing physicians, wherein the terms of dispensing medications are addressed including policies for monitoring compliance. Frequently, these policies involve submitting urine specimens for analysis. Antibody-based screening immunoassays are often employed, due to their low cost, rapid turnaround time and ease of use. Although considered best practice, not all presumptive positives are confirmed by more sensitive and specific liquid chromatography tandem mass spectrometry (LC–MS/MS) methods. Both the Clinical and Laboratory Standards Institute and The National Academy of Clinical Biochemists have recently introducing new practice guidelines for therapeutic drug monitoring. One recommendation aimed at ensuring specimen validity will be to test all urine toxicology specimens for pH, creatinine and specific gravity. Specimens that fail an initial integrity test should not be sent for further analysis and the ordering provider should be notified (2, 3). Case Description A 54-year-old Caucasian woman presented to the Pharmacy Pain Medication Therapy Management (MTM) Clinic at UF Health Jacksonville for prescription pickup. Patient’s past medical history was significant for bipolar disorder, degeneration of lumbar intervertebral disk, hypertension, hyperlipidemia and type 2 diabetes. Patient was currently taking hydrocodone/acetaminophen 7.5/325 mg one tablet every 6 h as needed, along with gabapentin, tizanidine and duloxetine. A random Urine Drugs of Abuse (DOA) assay was performed as a part of the controlled substance agreement. The medical assistant alerted the pharmacist when it was suspected that the patient submitted a highly diluted urine sample based on a visual inspection. At the time, the pharmacist ordered a urine creatinine, a serum DOA and contacted the laboratory to inform them of a potentially adulterated sample. Case Resolution and Discussion The patient’s previous drug screenings were noted positive for benzodiazepines and opiates, consistent with prescriptions for diazepam (Valium 10 mg q 12 h) and hydrocodone–acetaminophen (Norco 7.5–325 mg q 6 h). The laboratory employs qualitative versions of the kinetic interaction of microparticles in solution (KIMS)-based immunoassays (Roche Diagnostics). The assay cut-off values are 100 and 300 ng/mL for benzodiazepines and opiates, respectively, with manufacturer-reported cross-reactivity of 20% for hydrocodone and 94% for diazepam. The prescription strengths for hydrocodone had recently been reduced (10–7.5 mg), with diazepam discontinued a year prior. The patient had recently endorsed taking more hydrocodone than prescribed on multiple occasions, raising compliance concerns. The most recent urine specimen submitted was suspected to be heavily diluted at the time of collection based on appearance (clear; similar to water) (Figure 1D). This observation by the clinical team prompted contact with the laboratory for further investigation. Specimen validity testing was conducted, which revealed creatinine values below the limit of quantification (15 ng/mL) and specific gravity of 1.004. These results are characteristics of a highly dilute specimen and supported suspicions that the sample submitted was adulterated urine (Figure 1A and D). No previous creatinine values were noted for this patient, however previous urinalysis results revealed specific gravity values >1.02, consistent with normal urine. Figure 1. View largeDownload slide Patient’s urine specimen analytical and physical characteristics. (A,D) Specimen integrity testing and visual appearance (B) Roche KIMS drugs of abuse immunoassay panel results, positivity cut-off mAbs > 0 absorbance units. (C) Confirmatory analysis performed on urine specimen by LC–MS/MS. Figure 1. View largeDownload slide Patient’s urine specimen analytical and physical characteristics. (A,D) Specimen integrity testing and visual appearance (B) Roche KIMS drugs of abuse immunoassay panel results, positivity cut-off mAbs > 0 absorbance units. (C) Confirmatory analysis performed on urine specimen by LC–MS/MS. At the time of sample submission, the laboratory had yet to implement mandatory validity testing prior to sample analysis, therefore a full urine DOA panel was performed (Figure 1B). The sample screened positive for benzodiazepines, opiates and oxycodone (Roche Diagnostics, oxycodone cut-off 100 ng/mL). Because of the unexpected results of the specimen validity testing and immunoassay screen, the laboratory contacted the pharmacy team to obtain more information. Based on these results, we reasoned that this patient might be attempting to conceal polysubstance abuse by diluting the urine sample with water but was unsuccessful in creating a dilute enough specimen to avoid detection. We also reviewed the patient’s medication history against commonly interfering substances for the benzodiazepine, opiate and oxycodone assays, and none were noted in the electronic medical record (4). Sample dilution could account for the low positivity (absorbance values within 10% of the cut-off) noted for benzodiazepine and oxycodone and also the specimen’s general appearance. Upon closer inspection, we noted the sample to be cloudy, with sediment settled at the bottom of the cup (Figure 1A and D). To that end, sample adulteration by crushing or shaving off a part of a pill directly into the sample is noted in the literature as a method to deceive providers and laboratories (5, 6). The patient’s urine sample was subsequently sent out for confirmatory testing by LC–MS/MS. Interestingly, the sample tested negative for benzodiazepines (components of the testing included: alprazolam; clonazepam; flurazepam; lorazepam; midazolam; nordiazepam; oxazepam; temazepam; triazolam; with a cut-off of 100 ng/mL) and oxycodone (components of the testing included: oxycodone; oxymorphone with a cut-off of 100 ng/mL) but had very high levels of hydrocodone (4,700 ng/mL) and no detectable hydromorphone (Figure 1C). The reference laboratory (Laboratory Corporation of America) where the specimen was sent does not offer norhydrocodone testing as a part of their opiate confirmation panel. Hydrocodone undergoes metabolism by both CYP2D6 and CYP3A4 into hydromorphone and norhydrocodone, respectively (7). Another reason to explain the lack of detectable metabolites in the sample would include ethnicity. Of note, the patient is Caucasian and ~5–10% of Caucasians are considered poor metabolizers with no CYP2D6 enzyme activity (8). These patients would therefore not be able to generate hydromorphone. Additionally, some medications can inhibit CYP2D6. The patient was prescribed duloxetine at the time of the drug screen, which is a known inhibitor of CYP2D6 (9). Either of these factors could account for the lack of detectable hydromorphone in the urine specimen. At this point, the pharmacist contacted the patient regarding the results (highly dilute specimen with sediment settled at the bottom; confirmed hydrocodone and no metabolites), the patient at that point admitted to taking diazepam. When asked specifically about crushing a hydrocodone pill in the sample, the patient denied this. However, given the admitted benzodiazepine use, the patient had violated their agreement and was informed they would no longer to be eligible to receive controlled substances. In an attempt to determine the true nature of the substances the patient consumed, a serum sample obtained in parallel to the urine test was sent out to another lab for analysis by LC–MS/MS. Briefly, a serum calibrator prepared at a 10 ng/mL concentration of nordiazepam, temazepam and oxazepam along with hydrocodone, norhydrocodone, hydromorphone, oxycodone noroxycodone and oxymorphone was extracted along with the patient sample and analyzed via LC–MS for cut-off purposes. Interestingly, the results identified the presence of diazepam and common metabolites (nordiazepam, temazepam and oxazepam) (cut-off of 10 ng/mL). Hydrocodone and norhydrocodone were also present, however no hydromorphone or oxycodone (cut-off of 10 ng/mL) were detected. These results support the hypothesis that the patient took diazepam in addition to the prescribed hydrocodone and in an effort to conceal the diazepam use, diluted the urine and crushed in a hydrocodone pill. Of note, the high levels of opiate present in the sample due to the crushed pill are unlikely to result in a false-positive oxycodone result as the manufacturer lists no cross-reactivity with up to 75,000 ng/mL of hydrocodone. We attribute the low-positive urine benzodiazepine immunoassay screen to the sample being diluted however, we are unable to explain why when the sample was sent for confirmatory testing by LC–MS/MS with a lower cut-off (100 ng/mL vs 300 ng/mL) the sample came back negative. The serum testing however was consistent with the patients admitted diazepam use. This case illustrates the importance of active communication between clinical teams overseeing patient care and the laboratory in order to identify cases requiring a more scrutinizing look. Additionally, the use of specimen integrity measurements should be a standard part of any pain management program. References 1 Jones , C.M. , Paulozzi , L.J. , Mack , K.A. ( 2014 ) Sources of prescription opioid pain relievers by frequency of past-year nonmedical use United States, 2008–2011 . JAMA Internal Medicine , 174 , 802 – 803 . Google Scholar Crossref Search ADS PubMed 2 Jannetto , P.J. , Bratanow , N.C. , Clark , W.A. , Hamill-Ruth , R.J. , Hammett-Stabler , C.A. , Huestis , M.A. , et al. . ( 2018 ) Executive summary: American Association of Clinical Chemistry Laboratory Medicine Practice Guideline—using clinical laboratory tests to monitor drug therapy in pain management patients . The Journal of Applied Laboratory Medicine: An AACC Publication , 2 , 489 – 526 . Google Scholar Crossref Search ADS 3 Hammet-Stabler , C.A. , Eveline , A.S. , Brower , J. , Clarke , W. , Donovan , P. , Fitzgerald , R. , et al. . Laboratory Support for Pain Management Programs , 1st edition . Clinical and Laboratory Standards Institute guideline C63 : Wayne, PA , 2018 ; pp. 1 – 74 . 4 Saitman , A. , Park , H.D. , Fitzgerald , R.L. ( 2014 ) False-positive interferences of common urine drug screen immunoassays: a review . Journal of Analytical Toxicology , 38 , 387 – 396 . Google Scholar Crossref Search ADS PubMed 5 Pesce , A. , West , C. , Egan City , K. , Strickland , J. ( 2012 ) Interpretation of urine drug testing in pain patients . Pain Medicine , 13 , 868 – 885 . Google Scholar Crossref Search ADS PubMed 6 Lee , D. , Bazydlo , L.A. , Reisfield , G.M. , Goldberger , B.A. ( 2015 ) Urine spiking in a pain medicine clinic: an attempt to simulate adherence . Pain Medicine , 16 , 1449 – 1451 . Google Scholar Crossref Search ADS PubMed 7 Clarke , W. (ed). Contemporary Practice in Clinical Chemistry . American Association for Clinical Chemistry Press : Washington DC , 2016 ; p. 709 . 8 Alvan , G. , Bechtel , P. , Iselius , L. , Gundert-Remy , U. ( 1990 ) Hydroxylation polymorphisms of debrisoquine and mephenytoin in European populations . European Journal of Clinical Pharmacology , 39 , 533 – 537 . Google Scholar Crossref Search ADS PubMed 9 Skinner , M.H. , Kuan , H.Y. , Pan , A. , Sathirakul , K. , Knadler , M.P. , Gonzales , C.R. , et al. . ( 2003 ) Duloxetine is both an inhibitor and a substrate of cytochrome P4502D6 in healthy volunteers . Clinical Pharmacology and Therapeutics , 73 , 170 – 177 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Morrow, Jayne B; Ropero-Miller, Jeri D; Catlin, Megan L; Winokur, Agnes D; Cadwallader, Amy B; Staymates, Jessica L; Williams, Shannan R; McGrath, Jonathan G; Logan, Barry K; McCormick, Michael M; Nolte, Kurt B; Gilson, Thomas P; Menendez, M J; Goldberger, Bruce A

Williams,, Michelle;Martin,, Jennifer;Galettis,, Peter

doi: 10.1093/jat/bky043pmid: 30060217

Abstract Workplace drug testing in Australia is governed by two standards AS/NZS 4308:2008 for testing in urine and AS 4760:2006 for oral fluid. These standards are prescriptive and describe the drugs tested, procedures for analysis and collection devices. However, the drugs listed are not exhaustive and workers may consume novel psychoactive substances without detection. Here we present a validated method for the detection and quantitation of 19 synthetic cannabinoids in oral fluid. These drugs are AM2233, JWH-200, AB-005, AB-FUBINACA, AB-PINACA, AB-CHMINACA, AM2201, RCS-4, JWH-250, STS-135, JWH-73, XLR-11, JWH-251, JWH-18, JWH-122, JWH-19, UR-144, JWH-20 and AKB-48. The sample volume is 100 μL and is subject to a rapid, simple, protein precipitation step prior to centrifugation and injection into the LC–MS/MS system. Chromatographic separation was achieved in 4 min on a Kinetex Biphenyl column (50 mm × 3 mm × 2.6 μm) using 0.1% formic acid in water and acetonitrile as the mobile phase. The method was validated with a limit of detection (1 ng/mL) limit of quantitation (2.5 ng/mL), selectivity, linearity (2.5–500 ng/mL), accuracy (90.5–112.5% of the target concentration) and precision (3–14.7%). This method provides for the rapid detection of synthetic cannabinoids in oral fluid which is readily applicable to a routine laboratory. Introduction Synthetic cannabinoid is a broad term for the class of novel psychoactive substance (NPS) designed to mimic the effects of traditional cannabis. These compounds were first characterized through the legitimate research of John W Huffman at Clemson University yet remained largely unknown until recently (1). The evolution of synthetic cannabinoids has been in response to market pressure and legislation (2). The first compounds identified, while different in structure to Δ-9 Tetrahydrocannabinol (THC), were somewhat similar in that they contained an alkyl tail; some of these were JWH-73, JWH-18 and HU-210. These compounds pose a significant health risk as they bind more effectively to the CB1 and CB2 receptors (3) requiring a lower dose to produce similar effects to THC. The metabolites also have pharmacological activity greater than THC (4, 5). More recent variants of this drug class are structurally diverse, with some compounds such as AB-CHMINACA and AB-PINACA shown to be significantly more potent than THC (6). The addition of a halogen may also increase the selectivity for the CB1 receptor over CB2, depending on the moiety it replaces and location (7). Adverse events from exposure to these drugs range in severity from agitation and confusion to psychosis, seizures and tachycardia (8). Synthetic cannabinoids have also been identified as the cause or a contributing factor in a number of deaths (9). These compounds are sold in a pure crystalline form as research chemicals (10) or in herbal blends designed for smoking, vaping or used to produce a tea. These blends are labeled as incense, frequently carry the warning “not for human consumption” (11) and are packaged in bright packets with creative names. The unregulated nature of these products means there are no labeling requirements relating to the specific drug or dose. The variability in these products means that K2 purchased 1 week may not contain the same drug as K2 purchased the next. Furthermore, analysis has indicated that some packages contain multiple drugs (12) and that differences in the distribution of the drug on the plant material give rise to “hot spots” within the package (13). These drugs have been analyzed in blood, or fractions thereof (14, 15), urine (16, 17), oral fluid (18, 19) and hair (20). Each of these matrices has advantages such as the ease of collection of oral fluid or the long detection window of hair, along with some drawbacks, notably where urine is used, the detection of parent drug is unlikely and the availability of metabolite analytical standards is limited. However, the ultimate decision as to the preferred matrix, in workplace drug testing, remains with the requesting authority. Currently, there are two standards that govern workplace drug testing performed in Australia AS/NZS 4308:2008 for the detection of drugs of abuse in urine and AS 4760:2006 for the detection of drugs of abuse in oral fluid. Both of these standards prescribe the drug classes tested (opiates, amphetamine type substances, cocaine and metabolites and cannabis, with benzodiazepines only in urine) the procedures that must be adhered to and laboratory testing protocols. Oral fluid drug testing is often selected by an employer as it is preferred by industrial unions favoring the shorter detection window indicative of recent use, the collection does not require a secure toilet facility and can be collected under direct observation with minimal invasion of the donors’ privacy. Oral fluid testing also typically detects the parent drug, which may be deposited during smoking or insufflation as well as excreted into the oral fluid, rather than the drug metabolites detected when analyzing urine (21). As the pharmacokinetics of synthetic cannabinoids are largely unknown, this was assumed based upon the known profile of THC metabolism with extremely low concentrations of THC–COOH being detected in oral fluid. Furthermore, consistency with AS 4760 was sought which also specifies the parent drug (THC) compared to metabolite (THC–COOH) listed in AS/NZS 4308:2008. Here we present a validated method for the detection of 19 synthetic cannabinoids in oral fluid with minimal sample preparation and a rapid LC–MS-MS analysis. The synthetic cannabinoids included in this method are AM2233, JWH-200, AB-005, AB-FUBINACA, AB-PINACA, AB-CHMINACA, AM2201, RCS-4, JWH-250, STS-135, JWH-73, XLR-11, JWH-251, JWH-18, JWH-122, JWH-19, UR-144, JWH-20 and AKB-48. Table I outlines the structures of the drugs tested, their chemical names and where relevant, synonyms, the name in bold will be used throughout this paper. Table I. Synthetic cannabinoid structures and names Structure Common name and synonyms AM-2233 (2-iodophenyl)[1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl]-methanone JWH-200 [1-[2-(4-morpholinyl)ethyl]-1H-indol-3-yl]-1-naphthalenyl-methanone AB-005 [1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl](2,2,3,3-tetramethylcyclopropyl)-methanone AB-FUBINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-[(4-fluorophenyl)methyl]-1H-indazole-3-carboxamide AB-PINACA (S)-N-(1-amino-3-methyl-1-oxobutan-2-yl)-1-pentyl-1H-indazole-3-carboxamide AB-CHMINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-(cyclohexylmethyl)-1H-indazole-3-carboxamide AM-2201 [1-(5-fluoropentyl)-1H-indol-3-yl]-1-naphthalenyl-methanone RCS-4 (4-methoxyphenyl)(1-pentyl-1H-indol-3-yl)methanone BTM-4 E-4 OBT-199 JWH-250 1-(1-pentyl-1H-indol-3-yl)-2-(2-methoxyphenyl)-ethanone Structure Common name and synonyms AM-2233 (2-iodophenyl)[1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl]-methanone JWH-200 [1-[2-(4-morpholinyl)ethyl]-1H-indol-3-yl]-1-naphthalenyl-methanone AB-005 [1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl](2,2,3,3-tetramethylcyclopropyl)-methanone AB-FUBINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-[(4-fluorophenyl)methyl]-1H-indazole-3-carboxamide AB-PINACA (S)-N-(1-amino-3-methyl-1-oxobutan-2-yl)-1-pentyl-1H-indazole-3-carboxamide AB-CHMINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-(cyclohexylmethyl)-1H-indazole-3-carboxamide AM-2201 [1-(5-fluoropentyl)-1H-indol-3-yl]-1-naphthalenyl-methanone RCS-4 (4-methoxyphenyl)(1-pentyl-1H-indol-3-yl)methanone BTM-4 E-4 OBT-199 JWH-250 1-(1-pentyl-1H-indol-3-yl)-2-(2-methoxyphenyl)-ethanone Table I. Synthetic cannabinoid structures and names Structure Common name and synonyms AM-2233 (2-iodophenyl)[1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl]-methanone JWH-200 [1-[2-(4-morpholinyl)ethyl]-1H-indol-3-yl]-1-naphthalenyl-methanone AB-005 [1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl](2,2,3,3-tetramethylcyclopropyl)-methanone AB-FUBINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-[(4-fluorophenyl)methyl]-1H-indazole-3-carboxamide AB-PINACA (S)-N-(1-amino-3-methyl-1-oxobutan-2-yl)-1-pentyl-1H-indazole-3-carboxamide AB-CHMINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-(cyclohexylmethyl)-1H-indazole-3-carboxamide AM-2201 [1-(5-fluoropentyl)-1H-indol-3-yl]-1-naphthalenyl-methanone RCS-4 (4-methoxyphenyl)(1-pentyl-1H-indol-3-yl)methanone BTM-4 E-4 OBT-199 JWH-250 1-(1-pentyl-1H-indol-3-yl)-2-(2-methoxyphenyl)-ethanone Structure Common name and synonyms AM-2233 (2-iodophenyl)[1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl]-methanone JWH-200 [1-[2-(4-morpholinyl)ethyl]-1H-indol-3-yl]-1-naphthalenyl-methanone AB-005 [1-[(1-methyl-2-piperidinyl)methyl]-1H-indol-3-yl](2,2,3,3-tetramethylcyclopropyl)-methanone AB-FUBINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-[(4-fluorophenyl)methyl]-1H-indazole-3-carboxamide AB-PINACA (S)-N-(1-amino-3-methyl-1-oxobutan-2-yl)-1-pentyl-1H-indazole-3-carboxamide AB-CHMINACA N-[(1 S)-1-(aminocarbonyl)-2-methylpropyl]-1-(cyclohexylmethyl)-1H-indazole-3-carboxamide AM-2201 [1-(5-fluoropentyl)-1H-indol-3-yl]-1-naphthalenyl-methanone RCS-4 (4-methoxyphenyl)(1-pentyl-1H-indol-3-yl)methanone BTM-4 E-4 OBT-199 JWH-250 1-(1-pentyl-1H-indol-3-yl)-2-(2-methoxyphenyl)-ethanone The selection of drugs to include in this assay was difficult as there was no published data on the usage trends of specific drugs in Australia. The selection was based on the analytes listed in the Randox Toxicology Synthetic cannabinoids AB-PINACA, synthetic cannabinoids UR144/XLR11 and synthetic cannabinoids JWH- 250/RCS 8 ELISA kits. This list was refined further by eliminating the metabolites and the availability of reference materials in Australia. Methods Chemicals and reagents 1 mg/mL solutions of each AM2233, JWH-200, AB-005, AB-FUBINACA, AB-PINACA, AB-CHMINACA, AM2201, RCS-4, JWH-250, STS-135, JWH-73, XLR-11, JWH-251, JWH-18, JWH-122, JWH-19, UR-144, JWH-20 and AKB-48 were purchased from Lipomed (Arlesheim, Switzerland). JWH-250 d5, JWH-73 d7, JWH-18 d11 and JWH-122 d9 were purchased from Cayman Chemicals (Ann Arbour, MI, USA). Water was purified using a Merk Millipore, Milli-Q Advantage A10 system (Darmstadt, Germany). Reagent grade ≥95% formic acid and LC-MS grade Chromasolv® Acetonitrile were from Sigma-Aldrich (St Louis, MO, USA). Oral fluid samples Oral fluid samples were collected from laboratory staff by direct expectoration into a sterile tube, these samples were tipped into a larger vessel when pooled, leaving any sediment in the primary container for discard. The samples were not centrifuged prior to spiking. Preparation of internal standards, calibration solutions and quality controls A stock solution at 10 μg/mL of all analytes in ACN was prepared from the individual primary standards of 1 mg/mL. The calibration curve was generated by adding 50, 20, 10 or 5 μL stock solutions to 1 mL blank oral fluid producing the highest points on the calibrations curve (500, 200, 100 and 50 ng/mL, respectively). These were diluted further in blank oral fluid to produce the lower end of the calibration curve at 20, 10, 7.5, 5 and 2.5 ng/mL. High (300 ng/mL) and medium (30 ng/mL) QC solutions were prepared by adding 90 or 9 μL stock solution to 3 mL blank oral fluid. Low QC (3 ng/mL) was prepared by diluting 30 μL of the high QC in 3 mL of blank oral fluid. Internal standard solution was prepared by adding 20 μL of each JWH-18 d11, JWH-73 d7, JWH-122 d9 and JWH-250 d5 (100 μg/mL) to 100 mL of ACN giving a final concentration of 20 ng/mL. Sample preparation Samples were prepared by adding 100 μL spiked oral fluid to 300 μL water and 200 μL ACN containing the internal standards. The tubes were centrifuged at 5,000 r.p.m. (2,300 g) for 5 min. A 100 μL aliquot of the supernatant was transferred to autosampler vials and 1 μL was injected into an LC–MS-MS system. LC–MS-MS system The UHPLC system was a Shimadzu, Nexera X2 LC-30AD pumps, SIL-30 AC autosampler with a DGU-20A5 degassing unit and CTO-20A column oven (Kyoto, Japan). Chromatographic separation was performed on a Kinetix Biphenyl column (50 mm × 3 mm × 2.6 μm) purchased from Phenomenex (Torrence, CA, USA) held at 40°C. Formic acid 0.1% in water (A) and acetonitrile (B) were used as the mobile phases at a flow rate of 0.5 mL/min. The first stage of the chromatographic gradient is isocratic at 55%B for 1 min with a step up to 75%, then with a linear gradient increasing to 95%B at 4 min and holding for 1 min before returning to 55% at 5 min for 1 min column re-equilibration. The mass spectrometer was a 6,500 QTRAP (SCIEX, Framingham, MA, USA). The MS operated in electrospray positive mode with the following settings: Curtain gas—20, collision gas—medium. Ion spray voltage—5,500, source temperature—450°C, ion source gas—1–15 and ion source gas—2–20. Scheduled MRM mode was used for compound detection with a detection window set to 20 s around the expected retention time. Data acquisition was controlled by Analyst 1.6.3 and processed with MultiQuant 3.0 (SCIEX, Framingham, MA, USA). Validation of method Each analyte was optimized individually by direct injection into the MS via the inbuilt syringe drive. The parameters in Table II are those obtained from the compound optimization function in Analyst. Method validation was performed in accordance with National Association of Testing Authorities (NATA) guidelines (22), where limit of detection (LOD), limit of quantitation (LOQ), selectivity, linearity of calibration, precision, repeatability, ion suppression and measurement of uncertainty (MOU) were evaluated. Table II. Analyte MS parameters Analyte Precursion ion Product ion Retention time DP (volts) CE (volts) CXP (volts) Internal Std Ion ratio Ion supression % AM2233 458.9 112.1 0.43 91 27 12 JWH-18 d11 94.7 39.9 458.9 98.1 0.43 91 61 10 JWH-200 385.1 155.1 0.49 197 29 10 JWH-122 d9 43.1 29.46 385.1 114.1 0.49 197 31 12 AB-005 353.1 112.1 0.49 96 31 14 JWH-18 d11 63.1 24 353.1 125.1 0.49 96 29 14 AB-FUBINACA 368.8 253.1 0.8 51 33 16 JWH-73 d7 93.9 27.4 368.8 324.2 0.8 51 23 20 AB-PINACA 331.0 215.1 1.06 51 33 14 JWH-250 d5 96.9 26.4 331.0 286.2 1.06 51 21 20 AB-CHMINACA 357.1 241.1 1.38 46 37 14 JWH-122 d9 97.2 27.1 357.1 312.2 1.38 46 23 16 AM2201 360.0 232.1 1.96 176 33 16 JWH-250 d5 56.1 22.2 360.0 127.1 1.96 176 63 14 RCS-4 322.0 135.0 2.05 161 31 14 JWH-73 d7 20.1 21.4 322.0 77.1 2.05 161 69 8 JWH-250 336.0 121.1 2.13 156 25 14 JWH-250 d5 6.9 20.8 336.0 144.1 2.13 156 43 8 STS-135 383.1 135.1 2.17 31 39 8 JWH-250 d5 13.4 28 383.1 77.1 2.17 31 119 8 JWH-73 328.0 127.1 2.18 166 55 16 JWH-73 d7 45.8 20.5 328.0 155.1 2.18 166 31 12 XLR-11 330.1 125.2 2.23 171 29 8 JWH-73 d7 49.7 20.5 330.1 232.1 2.23 171 33 22 JWH-251 320.0 214.1 2.27 156 33 12 JWH-250 d5 89.8 14.2 320.0 105.1 2.27 156 31 12 JWH-18 342.0 155.1 2.41 171 33 10 JWH-18 d11 59.5 19.1 342.0 127.1 2.41 171 63 14 JWH-122 356.0 214.1 2.62 191 33 18 JWH-122 d9 72.1 16.3 356.0 141.1 2.62 191 51 8 JWH-19 356.0 155.1 2.66 191 33 8 JWH-18 d11 40.6 15.5 356.0 127.1 2.66 191 65 16 UR-144 311.9 125.1 2.76 156 25 12 JWH-73 d7 53.9 17.4 311.9 214.1 2.76 156 31 16 JWH-20 370.1 155.1 2.95 196 35 16 JWH-122 d9 45.8 31.9 370.1 127.1 2.95 196 67 12 AKB-48 366.0 135.1 3.31 136 51 14 JWH-250 d5 24.1 13.3 366.0 93.0 3.31 136 63 10 JWH-250 d5 341.0 121.0 2.11 126 27 14 JWH-73 d7 335.0 155.0 2.16 156 49 6 JWH-18 d11 353.0 155.0 2.37 100 33 16 JWH-122 d9 365.0 169.0 2.59 100 33 10 Analyte Precursion ion Product ion Retention time DP (volts) CE (volts) CXP (volts) Internal Std Ion ratio Ion supression % AM2233 458.9 112.1 0.43 91 27 12 JWH-18 d11 94.7 39.9 458.9 98.1 0.43 91 61 10 JWH-200 385.1 155.1 0.49 197 29 10 JWH-122 d9 43.1 29.46 385.1 114.1 0.49 197 31 12 AB-005 353.1 112.1 0.49 96 31 14 JWH-18 d11 63.1 24 353.1 125.1 0.49 96 29 14 AB-FUBINACA 368.8 253.1 0.8 51 33 16 JWH-73 d7 93.9 27.4 368.8 324.2 0.8 51 23 20 AB-PINACA 331.0 215.1 1.06 51 33 14 JWH-250 d5 96.9 26.4 331.0 286.2 1.06 51 21 20 AB-CHMINACA 357.1 241.1 1.38 46 37 14 JWH-122 d9 97.2 27.1 357.1 312.2 1.38 46 23 16 AM2201 360.0 232.1 1.96 176 33 16 JWH-250 d5 56.1 22.2 360.0 127.1 1.96 176 63 14 RCS-4 322.0 135.0 2.05 161 31 14 JWH-73 d7 20.1 21.4 322.0 77.1 2.05 161 69 8 JWH-250 336.0 121.1 2.13 156 25 14 JWH-250 d5 6.9 20.8 336.0 144.1 2.13 156 43 8 STS-135 383.1 135.1 2.17 31 39 8 JWH-250 d5 13.4 28 383.1 77.1 2.17 31 119 8 JWH-73 328.0 127.1 2.18 166 55 16 JWH-73 d7 45.8 20.5 328.0 155.1 2.18 166 31 12 XLR-11 330.1 125.2 2.23 171 29 8 JWH-73 d7 49.7 20.5 330.1 232.1 2.23 171 33 22 JWH-251 320.0 214.1 2.27 156 33 12 JWH-250 d5 89.8 14.2 320.0 105.1 2.27 156 31 12 JWH-18 342.0 155.1 2.41 171 33 10 JWH-18 d11 59.5 19.1 342.0 127.1 2.41 171 63 14 JWH-122 356.0 214.1 2.62 191 33 18 JWH-122 d9 72.1 16.3 356.0 141.1 2.62 191 51 8 JWH-19 356.0 155.1 2.66 191 33 8 JWH-18 d11 40.6 15.5 356.0 127.1 2.66 191 65 16 UR-144 311.9 125.1 2.76 156 25 12 JWH-73 d7 53.9 17.4 311.9 214.1 2.76 156 31 16 JWH-20 370.1 155.1 2.95 196 35 16 JWH-122 d9 45.8 31.9 370.1 127.1 2.95 196 67 12 AKB-48 366.0 135.1 3.31 136 51 14 JWH-250 d5 24.1 13.3 366.0 93.0 3.31 136 63 10 JWH-250 d5 341.0 121.0 2.11 126 27 14 JWH-73 d7 335.0 155.0 2.16 156 49 6 JWH-18 d11 353.0 155.0 2.37 100 33 16 JWH-122 d9 365.0 169.0 2.59 100 33 10 Table II. Analyte MS parameters Analyte Precursion ion Product ion Retention time DP (volts) CE (volts) CXP (volts) Internal Std Ion ratio Ion supression % AM2233 458.9 112.1 0.43 91 27 12 JWH-18 d11 94.7 39.9 458.9 98.1 0.43 91 61 10 JWH-200 385.1 155.1 0.49 197 29 10 JWH-122 d9 43.1 29.46 385.1 114.1 0.49 197 31 12 AB-005 353.1 112.1 0.49 96 31 14 JWH-18 d11 63.1 24 353.1 125.1 0.49 96 29 14 AB-FUBINACA 368.8 253.1 0.8 51 33 16 JWH-73 d7 93.9 27.4 368.8 324.2 0.8 51 23 20 AB-PINACA 331.0 215.1 1.06 51 33 14 JWH-250 d5 96.9 26.4 331.0 286.2 1.06 51 21 20 AB-CHMINACA 357.1 241.1 1.38 46 37 14 JWH-122 d9 97.2 27.1 357.1 312.2 1.38 46 23 16 AM2201 360.0 232.1 1.96 176 33 16 JWH-250 d5 56.1 22.2 360.0 127.1 1.96 176 63 14 RCS-4 322.0 135.0 2.05 161 31 14 JWH-73 d7 20.1 21.4 322.0 77.1 2.05 161 69 8 JWH-250 336.0 121.1 2.13 156 25 14 JWH-250 d5 6.9 20.8 336.0 144.1 2.13 156 43 8 STS-135 383.1 135.1 2.17 31 39 8 JWH-250 d5 13.4 28 383.1 77.1 2.17 31 119 8 JWH-73 328.0 127.1 2.18 166 55 16 JWH-73 d7 45.8 20.5 328.0 155.1 2.18 166 31 12 XLR-11 330.1 125.2 2.23 171 29 8 JWH-73 d7 49.7 20.5 330.1 232.1 2.23 171 33 22 JWH-251 320.0 214.1 2.27 156 33 12 JWH-250 d5 89.8 14.2 320.0 105.1 2.27 156 31 12 JWH-18 342.0 155.1 2.41 171 33 10 JWH-18 d11 59.5 19.1 342.0 127.1 2.41 171 63 14 JWH-122 356.0 214.1 2.62 191 33 18 JWH-122 d9 72.1 16.3 356.0 141.1 2.62 191 51 8 JWH-19 356.0 155.1 2.66 191 33 8 JWH-18 d11 40.6 15.5 356.0 127.1 2.66 191 65 16 UR-144 311.9 125.1 2.76 156 25 12 JWH-73 d7 53.9 17.4 311.9 214.1 2.76 156 31 16 JWH-20 370.1 155.1 2.95 196 35 16 JWH-122 d9 45.8 31.9 370.1 127.1 2.95 196 67 12 AKB-48 366.0 135.1 3.31 136 51 14 JWH-250 d5 24.1 13.3 366.0 93.0 3.31 136 63 10 JWH-250 d5 341.0 121.0 2.11 126 27 14 JWH-73 d7 335.0 155.0 2.16 156 49 6 JWH-18 d11 353.0 155.0 2.37 100 33 16 JWH-122 d9 365.0 169.0 2.59 100 33 10 Analyte Precursion ion Product ion Retention time DP (volts) CE (volts) CXP (volts) Internal Std Ion ratio Ion supression % AM2233 458.9 112.1 0.43 91 27 12 JWH-18 d11 94.7 39.9 458.9 98.1 0.43 91 61 10 JWH-200 385.1 155.1 0.49 197 29 10 JWH-122 d9 43.1 29.46 385.1 114.1 0.49 197 31 12 AB-005 353.1 112.1 0.49 96 31 14 JWH-18 d11 63.1 24 353.1 125.1 0.49 96 29 14 AB-FUBINACA 368.8 253.1 0.8 51 33 16 JWH-73 d7 93.9 27.4 368.8 324.2 0.8 51 23 20 AB-PINACA 331.0 215.1 1.06 51 33 14 JWH-250 d5 96.9 26.4 331.0 286.2 1.06 51 21 20 AB-CHMINACA 357.1 241.1 1.38 46 37 14 JWH-122 d9 97.2 27.1 357.1 312.2 1.38 46 23 16 AM2201 360.0 232.1 1.96 176 33 16 JWH-250 d5 56.1 22.2 360.0 127.1 1.96 176 63 14 RCS-4 322.0 135.0 2.05 161 31 14 JWH-73 d7 20.1 21.4 322.0 77.1 2.05 161 69 8 JWH-250 336.0 121.1 2.13 156 25 14 JWH-250 d5 6.9 20.8 336.0 144.1 2.13 156 43 8 STS-135 383.1 135.1 2.17 31 39 8 JWH-250 d5 13.4 28 383.1 77.1 2.17 31 119 8 JWH-73 328.0 127.1 2.18 166 55 16 JWH-73 d7 45.8 20.5 328.0 155.1 2.18 166 31 12 XLR-11 330.1 125.2 2.23 171 29 8 JWH-73 d7 49.7 20.5 330.1 232.1 2.23 171 33 22 JWH-251 320.0 214.1 2.27 156 33 12 JWH-250 d5 89.8 14.2 320.0 105.1 2.27 156 31 12 JWH-18 342.0 155.1 2.41 171 33 10 JWH-18 d11 59.5 19.1 342.0 127.1 2.41 171 63 14 JWH-122 356.0 214.1 2.62 191 33 18 JWH-122 d9 72.1 16.3 356.0 141.1 2.62 191 51 8 JWH-19 356.0 155.1 2.66 191 33 8 JWH-18 d11 40.6 15.5 356.0 127.1 2.66 191 65 16 UR-144 311.9 125.1 2.76 156 25 12 JWH-73 d7 53.9 17.4 311.9 214.1 2.76 156 31 16 JWH-20 370.1 155.1 2.95 196 35 16 JWH-122 d9 45.8 31.9 370.1 127.1 2.95 196 67 12 AKB-48 366.0 135.1 3.31 136 51 14 JWH-250 d5 24.1 13.3 366.0 93.0 3.31 136 63 10 JWH-250 d5 341.0 121.0 2.11 126 27 14 JWH-73 d7 335.0 155.0 2.16 156 49 6 JWH-18 d11 353.0 155.0 2.37 100 33 16 JWH-122 d9 365.0 169.0 2.59 100 33 10 Limit of detection was defined as the value where the analyte of interest could be identified with a signal to noise ratio greater than three, while having the appropriate retention time and ion ratio. Limit of quantitation was defined as the lowest value, where the analyte could be identified with a signal to noise ratio greater than 10 with adequate precision and accuracy (<20%CV (coefficient of variation) and ±20% target concentration). The LOQ was determined by analysis of seven replicates of the lowest concentration of the calibration curve. LOQ, LOD and linearity were determined around an administratively defined cutoff value of 5 ng/mL. Linearity was determined by the generation of four calibration curves from 2.5 to 500 ng/mL on four different days. The slope was plotted using the signal finder algorithm in SCIEX Multiquant Software. Imprecision, accuracy and measurement of uncertainty were determined by evaluating low (3 ng/mL), medium (30 ng/mL) and high (300 ng/mL) QC concentrations. Each was repeated three times on four different days with an additional seven on a fifth day for intraday precision. Imprecision was calculated by the %CV, and accuracy was calculated as the percent of the target concentration. CV was required to be within 15% (or 20% at LODQ) and accuracy was required to be within 85–115%. Instrumental repeatability was evaluated by injecting each QC three times and calculating the mean and standard deviation of the calculated value. MOU was calculated as two times this standard deviation. Carryover was assessed using a blank sample following each of the highest calibrators and QC samples; no carryover was detected. Ruggedness was not specifically evaluated though a number of parameters were changed during the routine preparation of samples. These included the order in which water, oral fluid and ACN were added, the temperature of the ACN internal standard and the time taken to centrifuge, up to 30 min. None of these parameters had any effect on the outcome of the test. Selectivity was evaluated by the assessment of interferences caused by endogenous and exogenous factors. Endogenous interferences were evaluated by investigating blank oral fluid collected from laboratory staff (n = 7) by direct expectoration into a tube. Exogenous inferences were investigated by the addition of 500 ng/mL drugs to blank oral fluid. The drugs tested were cathinone, ephedrone, methylone, flephedrone, 3,4–Methylenedioxyamphetamine, para-Methoxyamphetamine, methedrone, 3,4,5–Trimethoxyamphetamine, Methylenedioxyamphetamine, butylone, mephedrone, 3,4–Methylenedioxyethylamphetamine, 4-Methylethcathinone, pentedrone, N-Methyl-1,3–Benzodioxolylbutanamine, 4-Methylthioamphetamine, α-Pyrrolidinovalerophenone, 1-(4-methylphenyl)-2-(1-pyrrolidinyl)-1-butanone, 4-Bromo-2,5-dimethoxyphenethylamine (2C-B),3,4-Methylenedioxy-pyrovalerone, dimethoxybromoamphetamine, 4-(ethylthio)-2,5- dimethoxy-benzeneethanamine (2C-T-2), 1-[3-(trifluoromethyl)phenyl]-piperazine, dihydrochloride, 4-ethyl-2,5-dimethoxy-α-methyl-benzeneethanamine, 2,5-dimethoxy-4-(propylthio)-benzeneethanamine (2C-T-7), naphyrone, 5,6-Methylenedioxy-2-aminoindane, 2-Fluoromethamphetamine, 2,5-Dimethoxyamphetamine, 2-(4-chloro-2,5-dimethoxyphenyl)-N-(2- methoxybenzyl)ethanamine (25C-NBOMe), 4-bromo-2,5-dimethoxy-N-[(2-methoxyphenyl)methyl]- benzeneethanamine (25B-NBOMe), 2,5-dimethoxy-N-[(2-methoxyphenyl)methyl]-4-[(1-methylethyl)thio]- benzeneethanamine (25T4-NBOMe), Δ-9-THC, cannabidiol and 3,4-dichloro- N-[[1-(dimethylamino)cyclohexyl]methyl]-benzamide (AH-7,921). Additionally, bias was evaluated by assessing the calculated concentration to the actual concentration over five runs with a minimum of three samples in each. The acceptable bias is ±20%. Ion suppression and enhancement were evaluated by comparing 10 post extraction spiked samples with six pure standards in ACN. Ion suppression was found to range from 13.3 to 39.9%. Results Chromatographic separation was achieved under 4 min with a total run time of 6 min including column re-equilibration time (Figure 1). The step from 55 to 75% B at 1 min improved chromatography and decreased run time compared to a continuous gradient from 55 to 95%B over 5 or 6 min. The %CV for retention time across four validation runs was less than 2%. Quantification of each analyte was based on the most prominent peak with a second qualifier transition monitored to ensure accurate identification. Table II outlines the Analyte, Q1 and Q3 mass, retention time and optimized parameters for each analyte, where the first transition is used for quantification. Figure 1. View largeDownload slide (A) Chromatogram of all analytes 1-AM 2233, 2-JWH-200, 3-AB-005, 4-AB-FUBINACA, 5- AB-PIANCA, 6-AB-CHMINACA, 7-AM 2201, 8-RCS-4, 9-JWH-250, 10-STS-135, 11-JWH-73, 12-XLR-11, 13-JWH-250, 14-JWH-18, 15-JWH-122, 16-JWH-19, 17-UR-144, 18-JWH-20, 19-AKB-48. Internal standards (not displayed) JWH-250 d5, JWH-73 d7, JWH-18 d11, JWH-122 d9 elute at 2.11 min, 2.16 min, 2.37 min, 2.59 min, respectively (B) chromatogram of all analytes at LOQ and (C) negative genuine oral fluid sample. Figure 1. View largeDownload slide (A) Chromatogram of all analytes 1-AM 2233, 2-JWH-200, 3-AB-005, 4-AB-FUBINACA, 5- AB-PIANCA, 6-AB-CHMINACA, 7-AM 2201, 8-RCS-4, 9-JWH-250, 10-STS-135, 11-JWH-73, 12-XLR-11, 13-JWH-250, 14-JWH-18, 15-JWH-122, 16-JWH-19, 17-UR-144, 18-JWH-20, 19-AKB-48. Internal standards (not displayed) JWH-250 d5, JWH-73 d7, JWH-18 d11, JWH-122 d9 elute at 2.11 min, 2.16 min, 2.37 min, 2.59 min, respectively (B) chromatogram of all analytes at LOQ and (C) negative genuine oral fluid sample. All analytes were found to have a LOD of 1 ng/mL and LOQ of 2.5 ng/mL. The calibration curve was linear from 2.5 to 500 ng/mL when using a 1/x regression not forced through zero and all calibration curves had an R value greater than 0.97, no endogenous or exogenous interferences were detected. Interday, intraday precision and accuracy and SD values for the low (3), mid (30) and high (300) QCs are presented in Table III. Table III. Accuracy and precision parameters Intraday %CV n = 7 Interday %CV n = 12 Accuracy %CV n = 12 Mean ± SD LQC MQC HQC LQC MQC HQC LQC MQC HQC LQC MQC HQC STS-135 T1 383/135 3.7 5.0 10.8 14.7 5.2 5.7 112.5 102.6 99.2 3.38 ± 0.5 30.8 ± 1.6 297.6 ± 17.0 JWH-250 T1 336/121 4.2 6.2 10.6 3.5 5.5 9.8 102.4 102.2 97.5 3.07 ± 0.11 30.7 ± 1.6 292.4 ± 28.6 JWH-251 T1 320/214 2.7 6.8 9.1 5.6 2.3 3.9 103.7 101.3 97.2 3.11 ± 0.18 30.4 ± 0.7 291.5 ± 11.3 UR-144 T1 312/125 2.3 5.9 3.6 6.7 9.3 12.8 97.3 98.2 97.5 2.92 ± 0.20 29.4 ± 2.7 292.3 ± 37.4 RCS-4 T1 332/135 5.6 5.7 6.9 8.2 3.8 7.2 98.4 98.7 97.4 2.95 ± 0.24 29.6 ± 1.1 292.3 ± 21.0 JWH-73 T1 328/127 2.9 7.9 6.8 5.7 5.7 3.7 101.0 100.4 101.2 3.03 ± 0.17 30.1 ± 1.7 303.7 ± 11.3 XLR-11 T1 330/125 4.4 5.3 5.7 10.5 4.0 4.2 96.7 103.1 99.6 2.90 ± 0.30 30.9 ± 1.2 298.7 ± 12.7 JWH-18 T1 342/155 4.1 4.2 8.9 7.0 4.9 4.9 97.1 103.6 102.1 2.91 ± 0.20 31.1 ± 1.5 306.2 ± 15.1 AM2201 T1 360/232 3.4 6.9 6.0 4.3 5.8 3.7 100.2 101.1 101.1 3.10 ± 0.13 30.3 ± 1.8 303.2 ± 11.1 AKB-48 T1 366/135 1.8 4.7 6.9 4.3 2.3 5.7 98.4 99.6 102.9 2.95 ± 0.13 29.9 ± 0.7 308.6 ± 17.7 JWH-19 T1 356/155 4.3 5.3 5.8 11.0 6.0 9.5 92.9 98.4 100.8 2.79 ± 0.31 29.5 ± 1.8 302.4 ± 28.7 JWH-122 T2 356/214 3.8 3.1 6.7 7.3 2.9 9.0 98.2 99.1 103.2 2.95 ± 0.22 29.7 ± 0.9 309.5 ± 27.8 JWH-20 T2 370/155 3.6 5.5 7.0 11.6 4.9 12.0 90.5 101.6 95.2 2.72 ± 0.32 30.5 ± 1.5 286.7 ± 34.4 JWH-200 T1 285/155 8.1 6.6 9.0 10.9 6.8 4.9 103.9 104.3 101.0 3.12 ± 0.34 31.1 ± 2.1 303.1 ± 14.8 AB-CHMINACA T1 365/241 4.7 3.7 6.9 11.3 3.2 3.4 98.6 98.6 99.6 2.96 ± 0.34 29.6 ± 0.9 298.9 ± 10.2 AB-FUBINACA T1 369/235 9.8 5.2 7.4 7.4 5.4 3.2 97.7 102.1 100.3 2.93 ± 0.22 30.6 ± 1.7 301.0 ± 9.5 AB-PINACA T1 331/215 7.3 6.8 7.8 6.0 3.4 3.0 103.1 99.9 99.8 3.09 ± 0.18 30.0 ± 1.0 299.4 ± 8.9 AM2233 T2 458/112 8.8 5.6 10.8 11.7 11.1 6.6 98.1 94.4 103.9 2.94 ± 0.35 28.3 ± 3.1 311.7 ± 20.6 AB-005 T1 353/112 3.8 6.1 9.6 14.1 7.5 3.8 108.6 97.5 100.4 3.26 ± 0.46 29.3 ± 2.2 301.1 ± 11.4 Intraday %CV n = 7 Interday %CV n = 12 Accuracy %CV n = 12 Mean ± SD LQC MQC HQC LQC MQC HQC LQC MQC HQC LQC MQC HQC STS-135 T1 383/135 3.7 5.0 10.8 14.7 5.2 5.7 112.5 102.6 99.2 3.38 ± 0.5 30.8 ± 1.6 297.6 ± 17.0 JWH-250 T1 336/121 4.2 6.2 10.6 3.5 5.5 9.8 102.4 102.2 97.5 3.07 ± 0.11 30.7 ± 1.6 292.4 ± 28.6 JWH-251 T1 320/214 2.7 6.8 9.1 5.6 2.3 3.9 103.7 101.3 97.2 3.11 ± 0.18 30.4 ± 0.7 291.5 ± 11.3 UR-144 T1 312/125 2.3 5.9 3.6 6.7 9.3 12.8 97.3 98.2 97.5 2.92 ± 0.20 29.4 ± 2.7 292.3 ± 37.4 RCS-4 T1 332/135 5.6 5.7 6.9 8.2 3.8 7.2 98.4 98.7 97.4 2.95 ± 0.24 29.6 ± 1.1 292.3 ± 21.0 JWH-73 T1 328/127 2.9 7.9 6.8 5.7 5.7 3.7 101.0 100.4 101.2 3.03 ± 0.17 30.1 ± 1.7 303.7 ± 11.3 XLR-11 T1 330/125 4.4 5.3 5.7 10.5 4.0 4.2 96.7 103.1 99.6 2.90 ± 0.30 30.9 ± 1.2 298.7 ± 12.7 JWH-18 T1 342/155 4.1 4.2 8.9 7.0 4.9 4.9 97.1 103.6 102.1 2.91 ± 0.20 31.1 ± 1.5 306.2 ± 15.1 AM2201 T1 360/232 3.4 6.9 6.0 4.3 5.8 3.7 100.2 101.1 101.1 3.10 ± 0.13 30.3 ± 1.8 303.2 ± 11.1 AKB-48 T1 366/135 1.8 4.7 6.9 4.3 2.3 5.7 98.4 99.6 102.9 2.95 ± 0.13 29.9 ± 0.7 308.6 ± 17.7 JWH-19 T1 356/155 4.3 5.3 5.8 11.0 6.0 9.5 92.9 98.4 100.8 2.79 ± 0.31 29.5 ± 1.8 302.4 ± 28.7 JWH-122 T2 356/214 3.8 3.1 6.7 7.3 2.9 9.0 98.2 99.1 103.2 2.95 ± 0.22 29.7 ± 0.9 309.5 ± 27.8 JWH-20 T2 370/155 3.6 5.5 7.0 11.6 4.9 12.0 90.5 101.6 95.2 2.72 ± 0.32 30.5 ± 1.5 286.7 ± 34.4 JWH-200 T1 285/155 8.1 6.6 9.0 10.9 6.8 4.9 103.9 104.3 101.0 3.12 ± 0.34 31.1 ± 2.1 303.1 ± 14.8 AB-CHMINACA T1 365/241 4.7 3.7 6.9 11.3 3.2 3.4 98.6 98.6 99.6 2.96 ± 0.34 29.6 ± 0.9 298.9 ± 10.2 AB-FUBINACA T1 369/235 9.8 5.2 7.4 7.4 5.4 3.2 97.7 102.1 100.3 2.93 ± 0.22 30.6 ± 1.7 301.0 ± 9.5 AB-PINACA T1 331/215 7.3 6.8 7.8 6.0 3.4 3.0 103.1 99.9 99.8 3.09 ± 0.18 30.0 ± 1.0 299.4 ± 8.9 AM2233 T2 458/112 8.8 5.6 10.8 11.7 11.1 6.6 98.1 94.4 103.9 2.94 ± 0.35 28.3 ± 3.1 311.7 ± 20.6 AB-005 T1 353/112 3.8 6.1 9.6 14.1 7.5 3.8 108.6 97.5 100.4 3.26 ± 0.46 29.3 ± 2.2 301.1 ± 11.4 Table III. Accuracy and precision parameters Intraday %CV n = 7 Interday %CV n = 12 Accuracy %CV n = 12 Mean ± SD LQC MQC HQC LQC MQC HQC LQC MQC HQC LQC MQC HQC STS-135 T1 383/135 3.7 5.0 10.8 14.7 5.2 5.7 112.5 102.6 99.2 3.38 ± 0.5 30.8 ± 1.6 297.6 ± 17.0 JWH-250 T1 336/121 4.2 6.2 10.6 3.5 5.5 9.8 102.4 102.2 97.5 3.07 ± 0.11 30.7 ± 1.6 292.4 ± 28.6 JWH-251 T1 320/214 2.7 6.8 9.1 5.6 2.3 3.9 103.7 101.3 97.2 3.11 ± 0.18 30.4 ± 0.7 291.5 ± 11.3 UR-144 T1 312/125 2.3 5.9 3.6 6.7 9.3 12.8 97.3 98.2 97.5 2.92 ± 0.20 29.4 ± 2.7 292.3 ± 37.4 RCS-4 T1 332/135 5.6 5.7 6.9 8.2 3.8 7.2 98.4 98.7 97.4 2.95 ± 0.24 29.6 ± 1.1 292.3 ± 21.0 JWH-73 T1 328/127 2.9 7.9 6.8 5.7 5.7 3.7 101.0 100.4 101.2 3.03 ± 0.17 30.1 ± 1.7 303.7 ± 11.3 XLR-11 T1 330/125 4.4 5.3 5.7 10.5 4.0 4.2 96.7 103.1 99.6 2.90 ± 0.30 30.9 ± 1.2 298.7 ± 12.7 JWH-18 T1 342/155 4.1 4.2 8.9 7.0 4.9 4.9 97.1 103.6 102.1 2.91 ± 0.20 31.1 ± 1.5 306.2 ± 15.1 AM2201 T1 360/232 3.4 6.9 6.0 4.3 5.8 3.7 100.2 101.1 101.1 3.10 ± 0.13 30.3 ± 1.8 303.2 ± 11.1 AKB-48 T1 366/135 1.8 4.7 6.9 4.3 2.3 5.7 98.4 99.6 102.9 2.95 ± 0.13 29.9 ± 0.7 308.6 ± 17.7 JWH-19 T1 356/155 4.3 5.3 5.8 11.0 6.0 9.5 92.9 98.4 100.8 2.79 ± 0.31 29.5 ± 1.8 302.4 ± 28.7 JWH-122 T2 356/214 3.8 3.1 6.7 7.3 2.9 9.0 98.2 99.1 103.2 2.95 ± 0.22 29.7 ± 0.9 309.5 ± 27.8 JWH-20 T2 370/155 3.6 5.5 7.0 11.6 4.9 12.0 90.5 101.6 95.2 2.72 ± 0.32 30.5 ± 1.5 286.7 ± 34.4 JWH-200 T1 285/155 8.1 6.6 9.0 10.9 6.8 4.9 103.9 104.3 101.0 3.12 ± 0.34 31.1 ± 2.1 303.1 ± 14.8 AB-CHMINACA T1 365/241 4.7 3.7 6.9 11.3 3.2 3.4 98.6 98.6 99.6 2.96 ± 0.34 29.6 ± 0.9 298.9 ± 10.2 AB-FUBINACA T1 369/235 9.8 5.2 7.4 7.4 5.4 3.2 97.7 102.1 100.3 2.93 ± 0.22 30.6 ± 1.7 301.0 ± 9.5 AB-PINACA T1 331/215 7.3 6.8 7.8 6.0 3.4 3.0 103.1 99.9 99.8 3.09 ± 0.18 30.0 ± 1.0 299.4 ± 8.9 AM2233 T2 458/112 8.8 5.6 10.8 11.7 11.1 6.6 98.1 94.4 103.9 2.94 ± 0.35 28.3 ± 3.1 311.7 ± 20.6 AB-005 T1 353/112 3.8 6.1 9.6 14.1 7.5 3.8 108.6 97.5 100.4 3.26 ± 0.46 29.3 ± 2.2 301.1 ± 11.4 Intraday %CV n = 7 Interday %CV n = 12 Accuracy %CV n = 12 Mean ± SD LQC MQC HQC LQC MQC HQC LQC MQC HQC LQC MQC HQC STS-135 T1 383/135 3.7 5.0 10.8 14.7 5.2 5.7 112.5 102.6 99.2 3.38 ± 0.5 30.8 ± 1.6 297.6 ± 17.0 JWH-250 T1 336/121 4.2 6.2 10.6 3.5 5.5 9.8 102.4 102.2 97.5 3.07 ± 0.11 30.7 ± 1.6 292.4 ± 28.6 JWH-251 T1 320/214 2.7 6.8 9.1 5.6 2.3 3.9 103.7 101.3 97.2 3.11 ± 0.18 30.4 ± 0.7 291.5 ± 11.3 UR-144 T1 312/125 2.3 5.9 3.6 6.7 9.3 12.8 97.3 98.2 97.5 2.92 ± 0.20 29.4 ± 2.7 292.3 ± 37.4 RCS-4 T1 332/135 5.6 5.7 6.9 8.2 3.8 7.2 98.4 98.7 97.4 2.95 ± 0.24 29.6 ± 1.1 292.3 ± 21.0 JWH-73 T1 328/127 2.9 7.9 6.8 5.7 5.7 3.7 101.0 100.4 101.2 3.03 ± 0.17 30.1 ± 1.7 303.7 ± 11.3 XLR-11 T1 330/125 4.4 5.3 5.7 10.5 4.0 4.2 96.7 103.1 99.6 2.90 ± 0.30 30.9 ± 1.2 298.7 ± 12.7 JWH-18 T1 342/155 4.1 4.2 8.9 7.0 4.9 4.9 97.1 103.6 102.1 2.91 ± 0.20 31.1 ± 1.5 306.2 ± 15.1 AM2201 T1 360/232 3.4 6.9 6.0 4.3 5.8 3.7 100.2 101.1 101.1 3.10 ± 0.13 30.3 ± 1.8 303.2 ± 11.1 AKB-48 T1 366/135 1.8 4.7 6.9 4.3 2.3 5.7 98.4 99.6 102.9 2.95 ± 0.13 29.9 ± 0.7 308.6 ± 17.7 JWH-19 T1 356/155 4.3 5.3 5.8 11.0 6.0 9.5 92.9 98.4 100.8 2.79 ± 0.31 29.5 ± 1.8 302.4 ± 28.7 JWH-122 T2 356/214 3.8 3.1 6.7 7.3 2.9 9.0 98.2 99.1 103.2 2.95 ± 0.22 29.7 ± 0.9 309.5 ± 27.8 JWH-20 T2 370/155 3.6 5.5 7.0 11.6 4.9 12.0 90.5 101.6 95.2 2.72 ± 0.32 30.5 ± 1.5 286.7 ± 34.4 JWH-200 T1 285/155 8.1 6.6 9.0 10.9 6.8 4.9 103.9 104.3 101.0 3.12 ± 0.34 31.1 ± 2.1 303.1 ± 14.8 AB-CHMINACA T1 365/241 4.7 3.7 6.9 11.3 3.2 3.4 98.6 98.6 99.6 2.96 ± 0.34 29.6 ± 0.9 298.9 ± 10.2 AB-FUBINACA T1 369/235 9.8 5.2 7.4 7.4 5.4 3.2 97.7 102.1 100.3 2.93 ± 0.22 30.6 ± 1.7 301.0 ± 9.5 AB-PINACA T1 331/215 7.3 6.8 7.8 6.0 3.4 3.0 103.1 99.9 99.8 3.09 ± 0.18 30.0 ± 1.0 299.4 ± 8.9 AM2233 T2 458/112 8.8 5.6 10.8 11.7 11.1 6.6 98.1 94.4 103.9 2.94 ± 0.35 28.3 ± 3.1 311.7 ± 20.6 AB-005 T1 353/112 3.8 6.1 9.6 14.1 7.5 3.8 108.6 97.5 100.4 3.26 ± 0.46 29.3 ± 2.2 301.1 ± 11.4 Carryover was assessed using a blank sample following each of the highest calibrators and QC samples; no carryover was detected. Ruggedness was not specifically evaluated though a number of parameters were changed during the routine preparation of samples. These included the order in which water, oral fluid and ACN were added, the temperature of the ACN internal standard and the time taken to centrifuge, up to 30 min. None of these parameters had any effect on the outcome of the test. Discussion The method described in this study detects the presence of 19 synthetic cannabinoids in oral fluid, these drugs are AM2233, JWH-200, AB-005, AB-FUBINACA, AB-PINACA, AB-CHMINACA, AM2201, RCS-4, JWH-250, STS-135, JWH-73, XLR-11, JWH-251, JWH-18, JWH-122, JWH-19, UR-144, JWH-20 and AKB-48. Workplace drug testing in Australia is usually limited to the drugs listed in AS/NZS 4308:2008 or AS 4760:2006. As NPS are not prescribed under either standard, and testing options are limited, they are rarely tested for and thus detection rates are unknown though the resource sector first noted these drugs as a problem (23). This method has been specifically designed to meet the needs of workplaces wishing to test for this class of drug. The decision to use oral fluid as a matrix was based on two factors; policy and ease of development. A workplace is constrained in the development of a drug testing policy largely by industrial unions which view the sampling of urine as invasive to the donors' privacy both during collection and due to the longer detection window, particularly of cannabis. A workplace whose policy specifies oral fluid as the matrix will be unable to collect a urine sample however a workplace that uses urine routinely will have few issues implementing oral fluid testing as an additional test for this class. Second, analytically the detection of parent drug in urine is limited and rather than wait for analytical standards to become available we elected to detect the parent drug, more commonly found in oral fluid. This method offers a wide range of drugs, is rapid enough to limit the time a worker is stood down, can be adapted to include new drugs as they become available and will function as a deterrent to the continued use whilst at work. A brief literature search yielded eight published methods where synthetic cannabinoids have been detected in oral fluid (18, 19, 24–30). These methods are outlined in Table IV and have been comprehensively reviewed elsewhere (31). Two of the methods include more cannabinoids (19, 26) than the one presented here, however ours includes the more recent drugs such as AB-CHMINACA and STS-135. Six of the methods require significant sample preparation including SPE (24, 25, 29, 30) or a preparatory step as in followed by evaporation then reconstitution (19, 26). The remaining two methods use dilution only. All of the evaluated methods with one exception use a large sample volume. In these cases, where the test uses the entire sample produced, this would prevent reanalysis in the event of a challenge to the results or additional tests to be performed such as adulteration. The remaining method uses a small volume of sample and has a relatively short chromatographic run time, however only includes seven of the original synthetic cannabinoids (28). Table IV. Methods for the detection of synthetic cannabinoids in oral fluid Year published Cannabinoids Sample preparation Chromatographic run time (min) Sample volume LOD/LOQ Ref. 2011 JWH-18, JWH-73, JWH-250, CP47-497, CP47-497-C8, HU-210 SPE 12.2 inc re-equil 1 mL Quantisal LOQ 0.5 ng/mL (24) 2012 JWH-18, JWH-19, JWH-73, JWH-122, JWH-200, JWH-250, HU-210, CP47-497 AM 694, Nabilone 1:1 dilution with mobile phase A 9 250 μL LOD 1–20 ng/mL (18) 2013 JWH-18, JWH-73, JWH-200, JWH-250, HU-211, CP47-497, CP47-497-C8 SPE Pos mode 8Neg mode 6 500 μL LOD 0.025–1 ng/mL LOQ 0.1–2.5 ng/mL (25) 2013 AM-1220, AM-2201, AM-2233, AM-694, JWH-007, JWH-15, JWH-18, JWH-19, JWH-20, JWH-73, JWH-81, JWH-122, JWH-200, JWH-203, JWH-210, JWH-250, JWH-251, JWH-307, JWH-387, JWH-398, JWH-412, MAM-2201, Methanandamide, RCS-4, RCS-4 ortho isomer, RCS-8, WIN 48.098, WIN 55,212-2 Protein precipitation evaporation and reconstitution in 100 μL 12 200 μL LOD 0.02–0.4 ng/mL LOQ 0.2–4 ng/mL (26) Year published Cannabinoids Sample preparation Chromatographic run time (min) Sample volume LOD/LOQ Ref. 2011 JWH-18, JWH-73, JWH-250, CP47-497, CP47-497-C8, HU-210 SPE 12.2 inc re-equil 1 mL Quantisal LOQ 0.5 ng/mL (24) 2012 JWH-18, JWH-19, JWH-73, JWH-122, JWH-200, JWH-250, HU-210, CP47-497 AM 694, Nabilone 1:1 dilution with mobile phase A 9 250 μL LOD 1–20 ng/mL (18) 2013 JWH-18, JWH-73, JWH-200, JWH-250, HU-211, CP47-497, CP47-497-C8 SPE Pos mode 8Neg mode 6 500 μL LOD 0.025–1 ng/mL LOQ 0.1–2.5 ng/mL (25) 2013 AM-1220, AM-2201, AM-2233, AM-694, JWH-007, JWH-15, JWH-18, JWH-19, JWH-20, JWH-73, JWH-81, JWH-122, JWH-200, JWH-203, JWH-210, JWH-250, JWH-251, JWH-307, JWH-387, JWH-398, JWH-412, MAM-2201, Methanandamide, RCS-4, RCS-4 ortho isomer, RCS-8, WIN 48.098, WIN 55,212-2 Protein precipitation evaporation and reconstitution in 100 μL 12 200 μL LOD 0.02–0.4 ng/mL LOQ 0.2–4 ng/mL (26) Table IV. Methods for the detection of synthetic cannabinoids in oral fluid Year published Cannabinoids Sample preparation Chromatographic run time (min) Sample volume LOD/LOQ Ref. 2011 JWH-18, JWH-73, JWH-250, CP47-497, CP47-497-C8, HU-210 SPE 12.2 inc re-equil 1 mL Quantisal LOQ 0.5 ng/mL (24) 2012 JWH-18, JWH-19, JWH-73, JWH-122, JWH-200, JWH-250, HU-210, CP47-497 AM 694, Nabilone 1:1 dilution with mobile phase A 9 250 μL LOD 1–20 ng/mL (18) 2013 JWH-18, JWH-73, JWH-200, JWH-250, HU-211, CP47-497, CP47-497-C8 SPE Pos mode 8Neg mode 6 500 μL LOD 0.025–1 ng/mL LOQ 0.1–2.5 ng/mL (25) 2013 AM-1220, AM-2201, AM-2233, AM-694, JWH-007, JWH-15, JWH-18, JWH-19, JWH-20, JWH-73, JWH-81, JWH-122, JWH-200, JWH-203, JWH-210, JWH-250, JWH-251, JWH-307, JWH-387, JWH-398, JWH-412, MAM-2201, Methanandamide, RCS-4, RCS-4 ortho isomer, RCS-8, WIN 48.098, WIN 55,212-2 Protein precipitation evaporation and reconstitution in 100 μL 12 200 μL LOD 0.02–0.4 ng/mL LOQ 0.2–4 ng/mL (26) Year published Cannabinoids Sample preparation Chromatographic run time (min) Sample volume LOD/LOQ Ref. 2011 JWH-18, JWH-73, JWH-250, CP47-497, CP47-497-C8, HU-210 SPE 12.2 inc re-equil 1 mL Quantisal LOQ 0.5 ng/mL (24) 2012 JWH-18, JWH-19, JWH-73, JWH-122, JWH-200, JWH-250, HU-210, CP47-497 AM 694, Nabilone 1:1 dilution with mobile phase A 9 250 μL LOD 1–20 ng/mL (18) 2013 JWH-18, JWH-73, JWH-200, JWH-250, HU-211, CP47-497, CP47-497-C8 SPE Pos mode 8Neg mode 6 500 μL LOD 0.025–1 ng/mL LOQ 0.1–2.5 ng/mL (25) 2013 AM-1220, AM-2201, AM-2233, AM-694, JWH-007, JWH-15, JWH-18, JWH-19, JWH-20, JWH-73, JWH-81, JWH-122, JWH-200, JWH-203, JWH-210, JWH-250, JWH-251, JWH-307, JWH-387, JWH-398, JWH-412, MAM-2201, Methanandamide, RCS-4, RCS-4 ortho isomer, RCS-8, WIN 48.098, WIN 55,212-2 Protein precipitation evaporation and reconstitution in 100 μL 12 200 μL LOD 0.02–0.4 ng/mL LOQ 0.2–4 ng/mL (26) We believe this method has a number of advantages over tests that are currently available due to the number of compounds detected, the speed of the run, minimal sample volume required and minimal sample preparation. Range of compounds detected One limitation of this method is that it is MRM data acquisition as opposed to high-resolution MS. Therefore, it will require regular updates to maintain currency with new drugs discovered in the market. However, it does contain a number of the well-known “original” synthetic cannabinoids such as JWH-18 and JWH-73 as well as more recent compounds such as AB-CHMINACA. The prevalence of these drugs in the workplace is not known however a number of ELISA assays would be required to cover this range of drugs. Multiple tests would bring with it the issue of sample volume as the typical volume collected is less than 1 mL. Minimal sample volume and preparation Oral fluid is the preferred matrix for industrial unions as it can be collected under direct observation, while still maintaining the privacy of the donor. It is also more indicative of recent drug use and by extension impairment, which is the primary concern in the workplace. As discussed above, the volume collected is minimal, particularly among donors who have dry mouth or use stimulant drugs including caffeine, nicotine and pseudoephedrine. Further requirements for separate testing and referee samples as well as enough sample specimen to perform screening and confirmatory tests and the proposed addition of adulteration tests mean that all analyses must be completed using small aliquots of original sample, which is usually less than 1 mL. This method uses 100 μL of neat oral fluid which is diluted and subject to a protein precipitation step. There is no SPE, or concentration making the total preparation time for a single sample around 7 min including a 5-min centrifugation step. From this, 1 μL is injected into the column, thus the actual volume of neat oral fluid that could contaminate the column and mass spectrometer is very small. This, combined with the use of a guard column has produced consistent results in over 500 injections to date. Speed and adaptability and speed of run The run time of 6 min makes it useful in a high throughput laboratory. Because it is a relatively simple gradient, additional drugs can be easily added. Prior to publication, this assay has been setup in a commercial laboratory virtually unchanged and includes over 30 drugs though NATA accreditation is attained will not be applied to routine samples. Assay design note This assay was designed to function as a confirmatory test in the same manner as tests undertaken as part of AS 4760:2006 or AS/NZS 4308:2008 using a proposed cutoff value of 5 ng/mL. To meet the requirements of this standard, this would require QC concentrations to be within ±50% of this value and a typical ULOQ of 100 ng/mL. There is little known about the expected concentrations in genuine samples, the decision was made to make the working range as wide as possible and to space the QCs along the range. It is expected that this method will act as a starting point for entities wishing to undertake this testing to begin from and adapt to their particular regulatory framework. It is also expected that laboratories which use different techniques for data acquisition such as SWATH would need to validate the method according to their requirements. Limits of study Stability of novel psychoactive substances is poorly characterized and did not form part of this study. However, the stability of cathinones and synthetic cannabinoids was investigated following validation of these methods. This work investigated the stability and recovery of NPS in glass tubes with neat oral fluid as well as using a buffered system both with and without the collection pad. These samples were evaluated at a number of time points throughout the month and in three temperature conditions (room temperature, refrigerated and frozen) at three concentrations (H = 300 ng/mL, M = 30 ng/mL L = 3 ng/mL). Overall synthetic cannabinoids are most stable when stored in glass tubes (without buffer) either refrigerated or frozen. This method was applied to 12 authentic samples previously submitted for routine testing, however no cannabinoids were detected. By conducting the validation in authentic oral fluid, we believe the method produced is more robust than had it relied on synthetic oral fluid, however the lack of confirmed positive samples for comparison or an external quality assurance program is a challenge The step from 55 to 75%B at 1 min allows for a faster run time, however, should any compounds be added that elute during this rapid change in the gradient caution should be exercised to ensure the repeatability of the chromatography. Conclusion This method details the validation for the detection of 19 synthetic cannabinoids in oral fluid. The drugs are AM2233, JWH-200, AB-005, AB-FUBINACA, AB-PINACA, AB-CHMINACA, AM2201, RCS-4, JWH-250, STS-135, JWH-73, XLR-11, JWH-251, JWH-18, JWH-122, JWH-19, UR-144, JWH-20 and AKB-48. The major advantage of this method is the rapid run time, minimal sample volume and preparation and the ability to add new drugs quickly as the standards become available. References 1 Wiley , J.L. , Marusich , J.A. , Huffman , J.W. , Balster , R.L. , Thomas , B.F. ( 2011 ) Hijacking of basic research: the case of synthetic cannabinoids . Methods Report (RTI Press) , 2011 , 17971 . Google Scholar PubMed 2 Dargan , P.I. , Hudson , S. , Ramsey , J. , Wood , D.M. ( 2011 ) The impact of changes in UK classification of the synthetic cannabinoid receptor agonists in ‘Spice’ . The International Journal on Drug Policy , 22 , 274 – 277 . Google Scholar Crossref Search ADS PubMed 3 Evren , C. , Bozkurt , M. ( 2013 ) Synthetic cannabinoids: crisis of the decade . Dusunen Adam: Journal of Psychiatry & Neurological Sciences , 26 , 1 – 11 . 4 Rajasekaran , M. , Brents , L.K. , Franks , L.N. , Moran , J.H. , Prather , P.L. ( 2013 ) Human metabolites of synthetic cannabinoids JWH-018 and JWH-073 bind with high affinity and act as potent agonists at cannabinoid type-2 receptors . Toxicology and Applied Pharmacology , 269 , 100 – 108 . Google Scholar Crossref Search ADS PubMed 5 Brents , L.K. , Reichard , E.E. , Zimmerman , S.M. , Moran , J.H. , Fantegrossi , W.E. , Prather , P.L. ( 2011 ) Phase I hydroxylated metabolites of the K2 synthetic cannabinoid JWH-018 retain in vitro and in vivo cannabinoid 1 receptor affinity and activity . PLoS One , 6 , 1 – 9 . Google Scholar Crossref Search ADS 6 Wiley , J.L. , Marusich , J.A. , Lefever , T.W. , Antonazzo , K.R. , Wallgren , M.T. , Cortes , R.A. , et al. . ( 2015 ) AB-CHMINACA, AB-PINACA, and FUBIMINA: affinity and potency of novel synthetic cannabinoids in producing Δ9-Tetrahydrocannabinol–like effects in mice . The Journal of Pharmacology and Experimental Therapeutics , 354 , 328 . Google Scholar Crossref Search ADS PubMed 7 Huffman , J.W. , Szklennik , P.V. , Almond , A. , Bushell , K. , Selley , D.E. , He , H. , et al. . ( 2005 ) 1-Pentyl-3-phenylacetylindoles, a new class of cannabimimetic indoles . Bioorganic & Medicinal Chemistry Letters , 15 , 4110 – 4113 . Google Scholar Crossref Search ADS PubMed 8 Seely , K.A. , Lapoint , J. , Moran , J.H. , Fattore , L. ( 2012 ) Spice drugs are more than harmless herbal blends: a review of the pharmacology and toxicology of synthetic cannabinoids . Progress in Neuro-Psychopharmacology and Biological Psychiatry , 39 , 234 – 243 . Google Scholar Crossref Search ADS PubMed 9 Trecki , J. , Gerona , R.R. , Schwartz , M.D. ( 2015 ) Synthetic cannabinoid-related illnesses and deaths . The New England Journal of Medicine , 373 , 103 – 107 . Google Scholar Crossref Search ADS PubMed 10 Kneisel , S. , Bisel , P. , Brecht , V. , Broecker , S. , Müller , M. , Auwärter , V. ( 2012 ) Identification of the cannabimimetic AM-1220 and its azepane isomer (N-methylazepan-3-yl)-3-(1-naphthoyl)indole in a research chemical and several herbal mixtures . Forensic Toxicology , 30 , 126 – 134 . Google Scholar Crossref Search ADS 11 Penn , H.J. , Langman , L.J. , Unold , D. , Shields , J. , Nichols , J.H. ( 2011 ) Detection of synthetic cannabinoids in herbal incense products . Clinical Biochemistry , 44 , 1163 – 1165 . Google Scholar Crossref Search ADS PubMed 12 Langer , N. , Lindigkeit , R. , Schiebel , H.-M. , Papke , U. , Ernst , L. , Beuerle , T. ( 2016 ) Identification and quantification of synthetic cannabinoids in “spice-like” herbal mixtures: Update of the German situation for the spring of 2016 . Forensic Science International , 269 , 31 – 41 . Google Scholar Crossref Search ADS PubMed 13 Hillebrand , J. , Olszewski , D. , Sedefov , R. ( 2010 ) Legal highs on the Internet . Substance Use & Misuse , 45 , 330 – 340 . Google Scholar Crossref Search ADS PubMed 14 Kronstrand , R. , Roman , M. , Andersson , M. , Eklund , A. ( 2013 ) Toxicological findings of synthetic cannabinoids in recreational users . Journal of Analytical Toxicology , 37 , 534 – 541 . Google Scholar Crossref Search ADS PubMed 15 Guale , F. , Shahreza , S. , Walterscheid , J.P. , Chen , H.-H. , Arndt , C. , Kelly , A.T. , et al. . ( 2013 ) Validation of LC–TOF-MS screening for drugs, metabolites, and collateral compounds in forensic toxicology specimens . Journal of Analytical Toxicology , 37 , 17 – 24 . Google Scholar Crossref Search ADS PubMed 16 Beuck , S. , Möller , I. , Thomas , A. , Klose , A. , Schlörer , N. , Schänzer , W. , et al. . ( 2011 ) Structure characterisation of urinary metabolites of the cannabimimetic JWH-018 using chemically synthesised reference material for the support of LC-MS/MS-based drug testing . Analytical and Bioanalytical Chemistry , 401 , 493 – 505 . Google Scholar Crossref Search ADS PubMed 17 Scheidweiler , K.B. , Jarvis , M.J.Y. , Huestis , M.A. ( 2015 ) Nontargeted SWATH acquisition for identifying 47 synthetic cannabinoid metabolites in human urine by liquid chromatography-high-resolution tandem mass spectrometry . Analytical and Bioanalytical Chemistry , 407 , 883 – 897 . Google Scholar Crossref Search ADS PubMed 18 Strano-Rossi , S. , Anzillotti , L. , Castrignano , E. , Romolo , F.S. , Chiarotti , M. ( 2012 ) Ultra high performance liquid chromatography-electrospray ionization-tandem mass spectrometry screening method for direct analysis of designer drugs, “spice” and stimulants in oral fluid . Journal of Chromatography. A , 1258 , 37 – 42 . Google Scholar Crossref Search ADS PubMed 19 Kneisel , S. , Auwärter , V. , Kempf , J. ( 2013 ) Analysis of 30 synthetic cannabinoids in oral fluid using liquid chromatography-electrospray ionization tandem mass spectrometry . Drug Testing and Analysis , 5 , 657 – 669 . Google Scholar Crossref Search ADS PubMed 20 Salomone , A. , Luciano , C. , Di Corcia , D. , Gerace , E. , Vincenti , M. ( 2014 ) Hair analysis as a tool to evaluate the prevalence of synthetic cannabinoids in different populations of drug consumers . Drug Testing and Analysis , 6 , 126 – 134 . Google Scholar Crossref Search ADS PubMed 21 Allen , K.R. ( 2011 ) Screening for drugs of abuse: which matrix, oral fluid or urine? Annals of Clinical Biochemistry , 48 , 531 – 541 . Google Scholar Crossref Search ADS PubMed 22 NATA . ( 2013 ) Guidelines for the validation and verification of quantitative and qualitative test methods. Technical Note 17 October 2013 23 Davidson , S. ( 2014 ) Synthetic drug use rife in Australian Mines. https://www.miningnews.net/archive/news/1197462/synthetic-drug-rife-australian-mines . 24 Coulter , C. , Garnier , M. , Moore , C. ( 2011 ) Synthetic cannabinoids in oral fluid . Journal of Analytical Toxicology , 35 , 424 – 430 . Google Scholar Crossref Search ADS PubMed 25 De Castro , A. , Pineiro , B. , Lendoiro , E. , Cruz , A. , Lopez-Rivadulla , M. ( 2013 ) Quantification of selected synthetic cannabinoids and delta 9-tetrahydrocannabinol in oral fluid by liquid chromatography-tandem mass spectrometry . Journal of Chromatography A , 1295 , 99 – 106 . Google Scholar Crossref Search ADS PubMed 26 Kneisel , S. , Speck , M. , Moosmann , B. , Corneillie , T.M. , Butlin , N.G. , Auwarter , V. ( 2013 ) LC/ESI-MS/MS method for quantification of 28 synthetic cannabinoids in neat oral fluid and its application to preliminary studies on their detection windows . Analytical and Bioanalytical Chemistry , 405 , 4691 – 4706 . Google Scholar Crossref Search ADS PubMed 27 Kneisel , S. , Auwärter , V. ( 2012 ) Analysis of 30 synthetic cannabinoids in serum by liquid chromatography-electrospray ionization tandem mass spectrometry after liquid-liquid extraction . Journal of Mass Spectrometry: JMS , 47 , 825 – 835 . Google Scholar Crossref Search ADS PubMed 28 Rodrigues , W.C. , Catbagan , P. , Rana , S. , Wang , G. , Moore , C. ( 2013 ) Detection of synthetic cannabinoids in oral fluid using ELISA and LC-MS-MS . Journal of Analytical Toxicology , 37 , 526 – 533 . Google Scholar Crossref Search ADS PubMed 29 Amaratunga , P. , Thomas , C. , Lemberg , B.L. , Lemberg , D. ( 2014 ) Quantitative measurement of XLR11 and UR-144 in oral fluid by LC-MS-MS . Journal of Analytical Toxicology , 38 , 315 – 321 . Google Scholar Crossref Search ADS PubMed 30 Oiestad , E.L. , Johansen , U. , Christophersen , A.S. , Karinen , R. ( 2013 ) Screening of synthetic cannabinoids in preserved oral fluid by UPLC-MS/MS . Bioanalysis , 5 , 2257 – 2268 . Google Scholar Crossref Search ADS PubMed 31 Aldlgan , A.A. , Torrance , H.J. ( 2016 ) Bioanalytical methods for the determination of synthetic cannabinoids and metabolites in biological specimens . Trends in Analytical Chemistry , 80 , 444 – 457 . Google Scholar Crossref Search ADS © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Guerrieri,, Davide;Kjellqvist,, Fanny;Kronstrand,, Robert;Gréen,, Henrik

doi: 10.1093/jat/bky060pmid: 30215771

Abstract Every year new fentanyl analog compounds, or fentanyls, appear on the drug scene. Development of immunoassays dedicated for screening individual molecules is challenging due to the short-lived presence of these compounds on the recreational drug market. Therefore, we investigated the detecting capabilities of the immunalysis fentanyl direct enzyme-linked immunosorbent assay (ELISA) kit against fentanyl in whole blood, and determined the cross-reactivity of nine fentanyl analogs (2-fluorofentanyl, acetylfentanyl, acrylfentanyl, carfentanil, cyclopropylfentanyl, tetrahydrofuranylfentanyl, furanylfentanyl, ocfentanil, valerylfentanyl) to confirm its validity for the general screening of fentanyls. Immunalysis ELISA assay was used to test whole blood samples fortified with fentanyl on a TECAN Freedom EVOlyzer platform, according to manufacturer specifications. The kit successfully was validated for fentanyl screening with a cutoff set at 0.5 ng/mL, and all tested analogs, with the exclusion of carfentanil, were detected. The lowest cross-reactivity with the kit was obtained with furanylfentanyl (20% ± 1, 95% confidence intervals (CI)) and 4-fluoroisobutyrfentanyl (25% ± 1, 95% CI), while the highest was recorded using acetylfentanyl (99% ± 11, 95% CI) and acrylfentanyl (94% ± 10, 95% CI). Post-mortem samples containing fentanyl, acrylfentanyl, cyclopropylfentanyl, THF-fentanyl and 4-fluoroisobutyrfentanyl were screened, and sensitivity and specificity of each analog were calculated. Positive screening results were generated by all post-mortem cases containing fentanyl (n = 14), acrylfentanyl (n = 11), cyclopropylfentanyl (n = 14), tetrahydrofuranylfentanyl (n = 13) and 4-fluoroisobutyrfentanyl (n = 10). Concentration of post-mortem fentanyl samples ranged from 0.5 ng/mL (cutoff) to 230 ng/mL, while the range for analogs was 3.4–36 ng/mL (cyclopentylfentanyl), 0.76–370 ng/mL (4-fluoroisobutyrfentanyl), 0.02–12 ng/mL (acrylfentanyl) and 2–26 ng/mL (tetrahydrofuranylfentanyl). The immunalysis fentanyl direct ELISA kit was successfully validated and showed significant cross-reactivity for all tested fentanyls, except carfentanil, making it a suitable technique for fentanyl and fentanyl analogs screening. Introduction Opioids are responsible for the majority of drug-overdose associated fatal intoxications worldwide (1). Albeit limited in number when compared to heroin- and other prescription opioid-caused deaths (tramadol, oxycodone, hydrocodone, fentanyl), the danger posed by new synthetic opioids should not be overlooked (2). Of the new substances marketed by the end of 2016, 4% were novel synthetic opioids and, among these, the majority was comprised by analogs of the prescription drug fentanyl (2). Fentanyl is a synthetic opioid commonly prescribed for its potent anti-nociceptive activity; synthesized in the 1950s, its potency as μ-receptor agonist was shown to be 80–100 higher than morphine (3). Since 2014, the appearance of fentanyl analogs of unknown potency, toxicity and side effects, has brought increasing attention to this class of compounds. Such a rapidly increasing class creates a serious burden on the legislators, since the novelty of these molecular structures is exploited for circumventing the laws on scheduled substances (4). As compounds are being added every year to the list of scheduled drugs, new unscheduled substances are rapidly taking their place on the recreational drug scene. Indeed, the pool of already described analogs from which to select unscheduled molecules for introduction on the black market is vast—scientific literature described more than 200 of the ~1,400 existing fentanyl analogs (5). The continuous emerging of new compounds puts a notable strain on the organizations tasked with the identification and detection of such new molecules. In general, the use of enzyme-linked immunosorbent assay (ELISA) is common practice in forensic laboratories all over the world (6). Fentanyl test kits are available from multiple vendors, and have recently been shown to detect—to different degrees—the presence of various fentanyl analogs as well (7). The aims of the present work were to validate the immunalysis fentanyl direct ELISA assay in whole blood and to explore the possibilities of detecting other frequently encountered fentanyl analogs through cross-reactivity experiments. As a proof of concept, a number of authentic cases were analyzed and evaluated. Materials and methods The assay used in the validation and in the real case testing was fentanyl direct ELISA kit (Immunalysis KI-218-IMM). Negative and positive synthetic controls were provided by Immunalysis. For in-house controls and calibrator, the following standards were used: fentanyl, 1 mg/mL was purchased from Cerilliant, (TX, USA), 2-fluorofentanyl, 4-fluoroisobutyrfentanyl, acetylfentanyl, acrylfentanyl, cyclopropylfentanyl, tetrahydrofuranylfentanyl (THF-fentanyl); furanylfentanyl, ocfentanil and valerylfentanyl were purchased from Cayman Chemical Company (MI, USA). Carfentanil was purchased from Toronto Research Chemicals (ON, Canada) (Figure 1). Figure 1. View largeDownload slide Structure of fentanyl and fentanyl analogs. Figure 1. View largeDownload slide Structure of fentanyl and fentanyl analogs. Drug-free blood for in-house preparation of controls and calibrators was from donor blood batches containing citric acid monohydrate, sodium dihydrogen phosphate dihydrate, sodium citrate, glucose monohydrate, sodium, water for injection. Negative samples from in-house donors and post-mortem specimens contained sodium fluoride and sodium heparin. The assays were performed on a Freedom EVOlyzer 150 system (TECAN), loaded with TECAN-provided consumables. The performance of ELISA testing on authentic cases was evaluated using post-mortem samples containing fentanyl analogs; cases were selected after the presence of fentanyl or the analogs had been confirmed by liquid chromatography–mass spectrometry/mass spectrometry (LC–MS-MS). ELISA assay The assay was validated according to the EMEA Guideline on bioanalytical method validation (8) and SOFT/AAFS Forensic Laboratory Guidelines (9), and the method described by Schwope et al. (10) specifically for forensic ELISA screening on biological matrices The immunoassays were conducted according to manufacturer specifications. In brief, spiked blood or authentic case specimens were loaded onto the EVOlyzer, pre-diluted 1:10 in ddH2O by a predilution protocol by the machine and then used for the assay. A volume of 80 μL per sample was loaded on the assay plate, followed by 100 μL of immunalysis enzyme conjugate solution. After a 60-min incubation at room temperature in the dark, the wells were washed six times with 350 μL of ddH2O and then added with 100 μL of Immunalysis substrate reagent. After a 30-min incubation, 100 μL of stop solution were added to each well. The absorbance was measured at 450 nm and 620 nm by the reader-equipped Freedom EVOlyzer. Validation Validation consisted of two sections: Section 1 included limit of detection (LoD), linearity, imprecision (intraplate, interplate), accuracy, hook effect, plate drift and carryover; Section 2 included analytical specificity and sensitivity. Section 1 was performed with both unpooled and pooled blood, spiked on the day of the analysis. Section 2 was performed on post-mortem cases, and results were compared against LC–MS-MS confirmation analysis. Calculation of the results for each sample (B/B0) was as follows: sample raw optical density (B) normalized against the mean optical density of the negative controls on each plate (B0). Linearity Negative pooled blood was fortified with fentanyl to 19 different concentrations, ranging from 0 to 25 ng/mL. The samples were run in triplicates. The concentration range was selected to cover the concentrations measured in positive case samples at the Swedish National Board of Forensic Medicine (RMV) in 2015 and 2016. Limit of detection Negative blood from 18 different donors was analyzed. Nine negatives were from in-house donors, while nine were from purchased blood batches. The formula for calculating the LoD is defined as SignalLoD = Signalblank + 3σblank. The immunalysis ELISA kit is an inverted assay (lower concentrations result in higher signal intensity); therefore, the 3σblank.factor must be subtracted, rather than added, to set the stringiest criteria possible. The LoD concentration value was calculated by extrapolating the LoD value on the regression curve described in the linearity section. Intraplate imprecision and plate drift Negative pooled blood was fortified with fentanyl to five different concentrations (0, 0.3, 1.2, 4.7 and 18.8 ng/mL fentanyl). Each concentration was run on a random pattern on the plate (n = 17–18). Coefficient of variation was calculated for each concentration. Interplate imprecision Negative pooled blood was fortified with fentanyl to five different concentrations (0, 0.3, 1.2, 4.7 and 18.8 ng/mL fentanyl). Each concentration was run on different plates (n = 10). Coefficient of variation was calculated for each concentration as raw absorbance and normalized against the negative (B/B0). Hook effect Three samples with very high fentanyl concentrations, 100 ng/mL, were analyzed to rule out the presence of hook effect. Carryover To minimize the risk of carryover, disposable tips were used for pipetting the samples; washing routines were used in between each pipetting step, both for disposable tips and fixed needles. Nine negative samples were scattered throughout the plate in between positive samples. LC–MS-MS quantification Quantification of fentanyl and fentanyl analogs in femoral blood were performed as previously described and used by the National Board of Forensic Medicine for post-mortem identification of furanylfentanyl and acrylfentanyl (11, 12). Analysis was conducted on an LC-30AD liquid chromatography system, (Shimadzu Scientific Instruments, Kyoto, Japan) equipped with a Triple QuadTM 4500 System (AB SCIEX Instruments, Concord, Ontario) mass spectrometer mobile phase A (0.05% formic acid in 10 mM ammonium formate) and B (0.05% formic acid in methanol) were used at a flow rate set to 0.8 mL/min; the linear gradient was set from 2% B to 100% B in 3.0 min. An Acquity UPLC® BEH Phenyl (2.1 × 50 mm, 1.7 μm) (Waters, Milford, Massachusetts, USA) column was used, with an oven temperature set to 60°C. Electrospray in positive mode was used for ionization. Multiple reaction monitoring mode was used for data acquisition and specific for each compound. D5-fentanyl was used as internal standard (m/z 342.0/188.0). Dwell time was 20 ms. Data analysis was performed on the Analyst® 1.6.2 software (AB SCIEX Instruments, Concord, Ontario). Extraction: femoral whole blood (1 g) was extracted using 500 μL 1 M TRIS-buffer pH 11 and 3 mL tert-butylmethylether; after 10 min on horizontal shaker, samples were centrifuged for 10 min at 4,000 g and then frozen at −80˚C for 20 min. The collected organic phase was evaporated to dryness (40˚C, 5 psi in a TurboVap Evaporation System (Biotage)). For reconstitution 100 μL phase A (0.05% HFO in ACN, 50:50) were used, then samples were incubated for 15 min at RT. The injection volume was 5 μL. Fentanyl analogs analysis Pooled negative blood samples were fortified with fentanyl to five different concentrations (0, 0.3, 1.2, 4.7 and 18.8 ng/mL, Figure 2). Fortified samples were run as references on the same assay, along with samples fortified with the following analogs: 2-fluorofentanyl, acetylfentanyl, acrylfentanyl, carfentanil, cyclopropylfentanyl, THF-fentanyl, furanylfentanyl, ocfentanil and valerylfentanyl. Each analog was assayed at three concentration levels (0.3, 1.2, 4.7 ng/mL). All analogs were run in triplicate in two separate experiments. The selected concentration points fall below (one point) and above (two points) the cutoff in use for whole blood routine fentanyl screening at the RMV. Analog readings, plotted onto the fentanyl fortified-blood curve, were used to extrapolate cross-reactivity of each analog with respect to fentanyl at every concentration point. Absorbance values that did not meet LoD criteria were excluded from cross-reactivity calculation. According to the US pharmacopoeia guidelines, lack of parallelism should be assumed and similarity is to be proven (13). To prevent a possible miscalculation due to lack of parallelism of the elicited signals at different concentrations, cross-reactivity values with respect to fentanyl were calculated separately at each concentration point. Figure 2. View largeDownload slide Fentanyl standard curve for cross-reactivity extrapolating calculations (two replicates in triplicate). Figure 2. View largeDownload slide Fentanyl standard curve for cross-reactivity extrapolating calculations (two replicates in triplicate). Authentic cases analysis Post-mortem samples sensitivity and specificity Calculation of sensitivity and specificity was conducted on 11 negative post-mortem cases and 14 fentanyl-positive post-mortem cases; all cases were confirmed via LC–MS-MS analysis, and categorized as false positive (FP), false negative (FN), true positive (TP) or true negative samples (TN). In the LC–MS-MS method, confirmation cutoff was set at 0.05 ng/mL; samples with fentanyl concentration between ELISA cutoff (0.5 ng/mL) and LC–MS-MS cutoff (0.05 ng/mL) were considered negative for the purposes of the analysis. Parameters were calculated as follows: sensitivity: TP/(TP + FN); specificity: TN/(TN + FP). Confidence interval (CI) was set at 95%. Post-mortem analogs cases Forty-eight post-mortem samples, selected for the confirmed presence of fentanyl analogs, were run in singlicate for qualitative analysis. The same assay allocated duplicates of pooled in-house fortified cutoffs (0.5 ng/mL) and standards (0, 0.3, 1.2, 4.7 ng/mL, the lowest being below the selected cutoff and the other two above); the run included commercially provided positive and negative controls. Moreover, the fentanyl standard curve was used to extrapolate quantitative values for post-mortem drug concentration. Results and Discussion Validation Linearity The normalized signals of the fentanyl-spiked blood samples were plotted against concentration values on a [conc]–signal plot (Figure 3). Cubic regression was utilized for extrapolated calculations (Signal = −0.0942 × [logConc]3 + 0.2776 × [logConc]2 − 0.3824 × [logConc] + 0.3206). Figure 3. View largeDownload slide Concentration/signal plot of linearity validation (signal (B) is normalized against blank (B0)). Figure 3. View largeDownload slide Concentration/signal plot of linearity validation (signal (B) is normalized against blank (B0)). Limit of detection The LoD for drug-free-blood (n = 9) resulted to be 69.7% B/B0 (mean B/B0blank = 0.902, SD = 0.068), extrapolated to 0.21 ng/mL. The LoD for aggregated series of in-house donor blood and drug-free blood (n = 18) resulted to be 70.3% B/B0 (mean B/B0blank = 0.908, SD = 0.068), extrapolated to 0.20 ng/mL. No significant difference was detected between the two LoD. Intraplate imprecision and plate drift The % CV for the tested concentration levels all resulted below 15%, with the exception of 1.2 ng/mL fentanyl (17.5% CV). Complete values are reported in Table I. Table I. Fentanyl-spiked blood intraplate imprecision n = 17–18 Fentanyl (ng/mL) Abs SD %CV 0 2.902 0.20 6.9 0.3 1.615 0.22 13.6 1.2 0.801 0.14 17.5 4.7 0.424 0.04 9.4 18.8 0.251 0.02 8.0 Fentanyl (ng/mL) Abs SD %CV 0 2.902 0.20 6.9 0.3 1.615 0.22 13.6 1.2 0.801 0.14 17.5 4.7 0.424 0.04 9.4 18.8 0.251 0.02 8.0 Table I. Fentanyl-spiked blood intraplate imprecision n = 17–18 Fentanyl (ng/mL) Abs SD %CV 0 2.902 0.20 6.9 0.3 1.615 0.22 13.6 1.2 0.801 0.14 17.5 4.7 0.424 0.04 9.4 18.8 0.251 0.02 8.0 Fentanyl (ng/mL) Abs SD %CV 0 2.902 0.20 6.9 0.3 1.615 0.22 13.6 1.2 0.801 0.14 17.5 4.7 0.424 0.04 9.4 18.8 0.251 0.02 8.0 Interplate imprecision The % CV for the tested concentration levels all resulted below 15%, with the lowest CV at 0.3 ng/mL fentanyl (6.8% CV). Complete values are reported in Table II. Table II. Fentanyl-spiked blood interplate imprecision; CV values represent both raw absorbance (Abs) and normalized data (B/B0), n = 10 Fentanyl (ng/mL) Abs SD %CV Abs %CV B/B0 0 2.767 0.42 15.2 0 0.3 2.157 0.42 19.5 6.4 1.2 1.226 0.29 23.7 12.1 4.7 0.569 0.12 21.1 9.3 18.8 0.332 0.07 21.1 8.9 Fentanyl (ng/mL) Abs SD %CV Abs %CV B/B0 0 2.767 0.42 15.2 0 0.3 2.157 0.42 19.5 6.4 1.2 1.226 0.29 23.7 12.1 4.7 0.569 0.12 21.1 9.3 18.8 0.332 0.07 21.1 8.9 Table II. Fentanyl-spiked blood interplate imprecision; CV values represent both raw absorbance (Abs) and normalized data (B/B0), n = 10 Fentanyl (ng/mL) Abs SD %CV Abs %CV B/B0 0 2.767 0.42 15.2 0 0.3 2.157 0.42 19.5 6.4 1.2 1.226 0.29 23.7 12.1 4.7 0.569 0.12 21.1 9.3 18.8 0.332 0.07 21.1 8.9 Fentanyl (ng/mL) Abs SD %CV Abs %CV B/B0 0 2.767 0.42 15.2 0 0.3 2.157 0.42 19.5 6.4 1.2 1.226 0.29 23.7 12.1 4.7 0.569 0.12 21.1 9.3 18.8 0.332 0.07 21.1 8.9 Hook effect All high-concentration samples resulted in markedly positive results, ruling out the presence of a hook effect. Carryover All nine negative standards scattered among positive samples through the plate returned negative signals, ruling out the presence of carryover effects. Fentanyl analogs analysis All analogs, with the exclusion of carfentanil, showed detectable cross-reactivity at 4.7 ng/mL; furanylfentanyl and 2-fluorofentanyl showed the lowest cross-reactivity (20% ± 1 CI and 25% ± 3 CI, respectively), while acetylfentanyl and acrylfentanyl showed the highest (99% ± 24 CI and 94% ± 10 CI, respectively). Results are reported in Table III and plotted in Figures 4a and b. The dose–response curves of the analysis on analog-fortified blood (Figure 4a) present different slopes and profiles, suggesting a lack of parallelism between the measured effect of fentanyl and the tested analogs. Table III. Cross-reactivity of nine different fentanyl analogs compared to fentanyl (two replicates in triplicate) Fentanyl Acetylfentanyl Acrylfentanyl Cyclopropylfentanyl THF-Fentanyl ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 100 101 101 95 88 99 83 83 94 54 67 42 52 95% CI 0 1 1 2 7 24 1 3 10 5 1 3 5 Ocfentanil Valerylfentanyl 2-Fluorofentanyl Furanylfentanyl Carfentanil ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 38 32 36 49 40 25 20 95% CI 1 4 0 9 4 3 1 Fentanyl Acetylfentanyl Acrylfentanyl Cyclopropylfentanyl THF-Fentanyl ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 100 101 101 95 88 99 83 83 94 54 67 42 52 95% CI 0 1 1 2 7 24 1 3 10 5 1 3 5 Ocfentanil Valerylfentanyl 2-Fluorofentanyl Furanylfentanyl Carfentanil ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 38 32 36 49 40 25 20 95% CI 1 4 0 9 4 3 1 Table III. Cross-reactivity of nine different fentanyl analogs compared to fentanyl (two replicates in triplicate) Fentanyl Acetylfentanyl Acrylfentanyl Cyclopropylfentanyl THF-Fentanyl ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 100 101 101 95 88 99 83 83 94 54 67 42 52 95% CI 0 1 1 2 7 24 1 3 10 5 1 3 5 Ocfentanil Valerylfentanyl 2-Fluorofentanyl Furanylfentanyl Carfentanil ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 38 32 36 49 40 25 20 95% CI 1 4 0 9 4 3 1 Fentanyl Acetylfentanyl Acrylfentanyl Cyclopropylfentanyl THF-Fentanyl ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 100 101 101 95 88 99 83 83 94 54 67 42 52 95% CI 0 1 1 2 7 24 1 3 10 5 1 3 5 Ocfentanil Valerylfentanyl 2-Fluorofentanyl Furanylfentanyl Carfentanil ng/mL 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 0.3 1.2 4.7 Mean (%) cross-reactivity 38 32 36 49 40 25 20 95% CI 1 4 0 9 4 3 1 Figure 4. View largeDownload slide (a) Dose–response curves of the nine tested analogs, each graph shows fentanyl dose–response curve as reference. Signals that surpassed the LoD threshold were discarded (inverted assay); (b) cross-reactivity of analogs compared to fentanyl (%). Figure 4. View largeDownload slide (a) Dose–response curves of the nine tested analogs, each graph shows fentanyl dose–response curve as reference. Signals that surpassed the LoD threshold were discarded (inverted assay); (b) cross-reactivity of analogs compared to fentanyl (%). Authentic case analysis Post-mortem samples sensitivity and specificity All negative cases were confirmed being below the selected cutoff and in all positive results fentanyl presence was confirmed above the ELISA cutoff. For fentanyl, sensitivity = 100% with a 95% CI of 77–100%, and specificity = 100% with a 95% CI of 72%–100%. Data are reported in Table IV. Table IV. Sensitivity and specificity for fentanyl and four fentanyl analogs (post-mortem blood) n Sensitivity (%) CI (%) Specificity (%) CI (%) Accuracy (%) CI (%) Fentanyl 14 100 77 100 72 100 86 Cyclopropylfentanyl 14 100 77 100 72 100 86 4-Fluoroisobutyrfentanyl 10 100 69 100 72 100 84 Acrylfentanyl 11 100 72 100 72 100 85 THF-fentanyl 13 100 75 100 72 100 86 n Sensitivity (%) CI (%) Specificity (%) CI (%) Accuracy (%) CI (%) Fentanyl 14 100 77 100 72 100 86 Cyclopropylfentanyl 14 100 77 100 72 100 86 4-Fluoroisobutyrfentanyl 10 100 69 100 72 100 84 Acrylfentanyl 11 100 72 100 72 100 85 THF-fentanyl 13 100 75 100 72 100 86 Table IV. Sensitivity and specificity for fentanyl and four fentanyl analogs (post-mortem blood) n Sensitivity (%) CI (%) Specificity (%) CI (%) Accuracy (%) CI (%) Fentanyl 14 100 77 100 72 100 86 Cyclopropylfentanyl 14 100 77 100 72 100 86 4-Fluoroisobutyrfentanyl 10 100 69 100 72 100 84 Acrylfentanyl 11 100 72 100 72 100 85 THF-fentanyl 13 100 75 100 72 100 86 n Sensitivity (%) CI (%) Specificity (%) CI (%) Accuracy (%) CI (%) Fentanyl 14 100 77 100 72 100 86 Cyclopropylfentanyl 14 100 77 100 72 100 86 4-Fluoroisobutyrfentanyl 10 100 69 100 72 100 84 Acrylfentanyl 11 100 72 100 72 100 85 THF-fentanyl 13 100 75 100 72 100 86 Post-mortem analogs cases The results of the ELISA testing matched with the LC–MS-MS confirmation: of the post-mortem samples, all 48 analog-confirmed samples showed positive results in the ELISA screening. The quantitative results of the ELISA/LC–MS-MS comparison are detailed in Table V, while calculated sensitivity and specificity for the four tested analogs are reported in Table IV for comparison with fentanyl. Table V. Quantitative ELISA testing of LC–MS-MS confirmed post-mortem samples. Negative samples 10 and 11 showed fentanyl concentrations below the ELISA cutoff. (+: detected but not quantified) Analyte Method Sample # (ng/mL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Negative ELISA −0.1 −0.1 −0.1 0.3 −0.1 −0.1 −0.1 −0.1 −0.1 0.2 0.4 LC−MS-MS Neg Neg Neg Neg Neg Neg Neg Neg Neg 0.34 0.37 Fentanyl ELISA 1.5 7.5 1.2 1.7 2.0 2.3 2.3 2.7 4.7 4.6 5.5 5.5 5.8 6.0 LC–MS-MS 1.6 1.8 1.9 2 2.1 2.7 3.3 3.4 8.9 20 29 34 60 230 Cyclopropylfentanyl ELISA 6.8 8.2 9.6 8.4 8.5 8.2 9.4 13.1 12.4 11.5 13.2 12.2 13.7 12.6 LC–MS-MS 3.4 4.5 6.5 7.7 8 8.9 9.8 11 19 20 22 28 36 37 4-Fluoroisobutyrfentanyl ELISA 12.3 9.3 3.7 9.2 7.2 13.8 12.9 7.5 6.1 12.3 LC–MS-MS 0.8 6.6 14 31 32 44 71 78 101 370 Acrylfentanyl ELISA 11.2 7.4 15.6 13.6 15.4 8.7 12.1 13.3 17.0 19.6 14.8 LC–MS-MS 0.02 0.17 0.36 0.37 0.43 0.61 1.1 1.2 2.8 6.2 12.2 THF-fentanyl ELISA 9.7 12.1 3.0 16.0 18.1 14.3 14.1 13.4 8.9 13.6 12.6 11.8 9.3 LC–MS-MS 2 2.7 26 + + + + + + + + + + Analyte Method Sample # (ng/mL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Negative ELISA −0.1 −0.1 −0.1 0.3 −0.1 −0.1 −0.1 −0.1 −0.1 0.2 0.4 LC−MS-MS Neg Neg Neg Neg Neg Neg Neg Neg Neg 0.34 0.37 Fentanyl ELISA 1.5 7.5 1.2 1.7 2.0 2.3 2.3 2.7 4.7 4.6 5.5 5.5 5.8 6.0 LC–MS-MS 1.6 1.8 1.9 2 2.1 2.7 3.3 3.4 8.9 20 29 34 60 230 Cyclopropylfentanyl ELISA 6.8 8.2 9.6 8.4 8.5 8.2 9.4 13.1 12.4 11.5 13.2 12.2 13.7 12.6 LC–MS-MS 3.4 4.5 6.5 7.7 8 8.9 9.8 11 19 20 22 28 36 37 4-Fluoroisobutyrfentanyl ELISA 12.3 9.3 3.7 9.2 7.2 13.8 12.9 7.5 6.1 12.3 LC–MS-MS 0.8 6.6 14 31 32 44 71 78 101 370 Acrylfentanyl ELISA 11.2 7.4 15.6 13.6 15.4 8.7 12.1 13.3 17.0 19.6 14.8 LC–MS-MS 0.02 0.17 0.36 0.37 0.43 0.61 1.1 1.2 2.8 6.2 12.2 THF-fentanyl ELISA 9.7 12.1 3.0 16.0 18.1 14.3 14.1 13.4 8.9 13.6 12.6 11.8 9.3 LC–MS-MS 2 2.7 26 + + + + + + + + + + Table V. Quantitative ELISA testing of LC–MS-MS confirmed post-mortem samples. Negative samples 10 and 11 showed fentanyl concentrations below the ELISA cutoff. (+: detected but not quantified) Analyte Method Sample # (ng/mL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Negative ELISA −0.1 −0.1 −0.1 0.3 −0.1 −0.1 −0.1 −0.1 −0.1 0.2 0.4 LC−MS-MS Neg Neg Neg Neg Neg Neg Neg Neg Neg 0.34 0.37 Fentanyl ELISA 1.5 7.5 1.2 1.7 2.0 2.3 2.3 2.7 4.7 4.6 5.5 5.5 5.8 6.0 LC–MS-MS 1.6 1.8 1.9 2 2.1 2.7 3.3 3.4 8.9 20 29 34 60 230 Cyclopropylfentanyl ELISA 6.8 8.2 9.6 8.4 8.5 8.2 9.4 13.1 12.4 11.5 13.2 12.2 13.7 12.6 LC–MS-MS 3.4 4.5 6.5 7.7 8 8.9 9.8 11 19 20 22 28 36 37 4-Fluoroisobutyrfentanyl ELISA 12.3 9.3 3.7 9.2 7.2 13.8 12.9 7.5 6.1 12.3 LC–MS-MS 0.8 6.6 14 31 32 44 71 78 101 370 Acrylfentanyl ELISA 11.2 7.4 15.6 13.6 15.4 8.7 12.1 13.3 17.0 19.6 14.8 LC–MS-MS 0.02 0.17 0.36 0.37 0.43 0.61 1.1 1.2 2.8 6.2 12.2 THF-fentanyl ELISA 9.7 12.1 3.0 16.0 18.1 14.3 14.1 13.4 8.9 13.6 12.6 11.8 9.3 LC–MS-MS 2 2.7 26 + + + + + + + + + + Analyte Method Sample # (ng/mL) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Negative ELISA −0.1 −0.1 −0.1 0.3 −0.1 −0.1 −0.1 −0.1 −0.1 0.2 0.4 LC−MS-MS Neg Neg Neg Neg Neg Neg Neg Neg Neg 0.34 0.37 Fentanyl ELISA 1.5 7.5 1.2 1.7 2.0 2.3 2.3 2.7 4.7 4.6 5.5 5.5 5.8 6.0 LC–MS-MS 1.6 1.8 1.9 2 2.1 2.7 3.3 3.4 8.9 20 29 34 60 230 Cyclopropylfentanyl ELISA 6.8 8.2 9.6 8.4 8.5 8.2 9.4 13.1 12.4 11.5 13.2 12.2 13.7 12.6 LC–MS-MS 3.4 4.5 6.5 7.7 8 8.9 9.8 11 19 20 22 28 36 37 4-Fluoroisobutyrfentanyl ELISA 12.3 9.3 3.7 9.2 7.2 13.8 12.9 7.5 6.1 12.3 LC–MS-MS 0.8 6.6 14 31 32 44 71 78 101 370 Acrylfentanyl ELISA 11.2 7.4 15.6 13.6 15.4 8.7 12.1 13.3 17.0 19.6 14.8 LC–MS-MS 0.02 0.17 0.36 0.37 0.43 0.61 1.1 1.2 2.8 6.2 12.2 THF-fentanyl ELISA 9.7 12.1 3.0 16.0 18.1 14.3 14.1 13.4 8.9 13.6 12.6 11.8 9.3 LC–MS-MS 2 2.7 26 + + + + + + + + + + The present study focuses on the validation of the Fentanyl ELISA kit commercialized by immunalysis, and on the cross-reactivity showed by fentanyl analogs in both fortified blood and post-mortem cases. Upon completion of the validation, the tests on drug-free blood fortified with nine different fentanyl analogs showed remarkable detecting powers for eight of the tested compounds. The various moieties comprising the fentanyl scaffold are known to have specific effects in the interaction with the receptor, and a plethora of substitutions has been reported to affect the potency and efficacy of the opiates to a various degree (14). Substitutions on the piperidine ring are known to drastically affect the analgesic potency of fentanyl analogs (15), and are not present among the selected compounds—with the exception of carfentanil. The analogs tested in this work present substitutions on the aniline moiety—playing an important role in eliciting opiate effect (16)—and on the propionil moiety—which in turn is involved in the affinity to the various opioid receptor (17). Interestingly, compounds with a small, apolar substituent on the propionil moiety, such as acetyl- and acrylfentanyl present cross-reactivity levels comparable to fentanyl. The more the substituent on the propionyl moiety increases in size, and therefore in steric effect, the more the interaction with the antibody appears encumbered, resulting in a decreased cross-reactivity (in order of increasing substituent size: cyclopropylfentanyl, valerylfentanyl). Moreover, increased polarity of the substituents on the aniline ring also contributes to a reduction in cross-reactivity; e.g., from ocfentanil and THF-fentanyl, which still present around 40% cross-reactivity, down to 2-fluorofentanyl and furanylfentanyl, which present much lower values. Finally, the only piperidine-substituted compound, carfentanil, presents an additional carbomethoxy group on Position 4 of the piperidine ring; such substitution seems to completely negate the interaction with the antibody in the assay, despite being known to produce a striking increase in nociceptive activity (18). All the post-mortem samples, containing either fentanyl—above the cutoff level—or any of the four analyzed analogs, resulted positive in the ELISA screening, supporting the hypothesis that the structural similarities of the analogs could allow for analog detection in a fentanyl-dedicated assay. It is important to highlight that ELISA forensic routine screening operates qualitatively, relying on a second alternative method for confirmation and quantification—in the present case, LC–MS-MS. Therefore, the validation of the ELISA kit is optimized for maximizing linearity in the proximity of the desired cutoff, rather than for quantitative and semi-quantitative ELISA testing. Nonetheless, the comparison between ELISA and LC–MS-MS quantitative performance could be of interest, with the caveat that the accuracy of ELISA quantification is bound to rapidly decrease away from cutoff concentration levels. Interesting are the samples with the lowest six (and possibly eight) concentrations of acrylfentanyl as confirmed by LC–MS-MS: adjusting for the 1:1 predilution and for the cross-reactivity percentage, the concentration of analogs in the aforementioned samples should have fallen below the 0.5 ng/mL cutoff set for fentanyl; nonetheless, all samples returned a positive signal on the ELISA. A possible explanation could be cross-reactivity with the assay of major acrylfentanyl metabolites. As described by Watanabe et al. (2017) in in vitro human hepatocytes and in vivo urine specimens, major acrylfentanyl metabolites are the nor-metabolites, but also hydroxy-acrylfentanyl (8). According to the manufacturer datasheet, the assay has negligible cross-reactivity with the main metabolite norfentanyl, but significant cross-reactivity for hydroxy-fentanyl (83%). Given the high similarity between fentanyl and acrylfentanyl, it is conceivable that high cross-reactivity would be present also between hydroxy-acrylfentanyl and the antibody (9). It is been reported that positive ELISA fentanyl screening could be caused by the presence of cross-reacting analogs in the specimens (19, 20); moreover, other ELISA fentanyl kits (NEOGEN) have recently been shown to be effective tools for detecting cross-reacting fentanyl analogs (7). Indeed, given the high sensitivity of the Immunalysis Fentanyl direct ELISA kit, it is possible to detect the presence of analogs in post-mortem blood with various cross-reactivity: acrylfentanyl and cyclopropylfentanyl positive cases returned positive ELISA results, and even analogs with low cross-reactivity such as THF-fentanyl and 4-fluoroisobutyrfentanyl showed signals above the screening cutoff. It is interesting to compare the cross-reactivity of fentanyl analogs on both Neogen and Immunalysis kit: according to Tiscione and Wegner (2017) acetylfentanyl and furanylfentanyl showed 31% and 63% cross-reactivity in blood, respectively (7). As mentioned in this work, acetylfentanyl and furanylfentanyl behaved differently, with a cross-reactivity at 4.7 ng/mL in blood of 99% and 20%, respectively. Differences are mostly to be attributed to the specific characteristics of the anti-fentanyl antibodies, and to the fragments of the fentanyl molecule they have been raised against. Nonetheless, it is possible to suggest that such differences will result in each kit detecting a diverse range of analogs. Conclusions The Immunalysis Fentanyl ELISA kit is able to detect, with different success rate, various fentanyl analogs in post-mortem blood. Given the rapid turnover of such analogs on the market, a fentanyl-positive screening could indeed be confirmed negative for fentanyl; nonetheless, to prevent a wrong interpretation as FP cases, an extended analysis for possible analogs—especially for those with small, apolar substituents—is recommended. Funding This research was funded by the Strategic Research Area in Forensic Sciences (Strategiområdet Forensiska Vetenskaper) at Linköping University and the National Board of Forensic Medicine. References 1 UNODC . ( 2017 ). Global overview of drug demand and supply. World Drug Report 2. 2 UNDOC . ( 2017 ). Market analysis of synthetic drugs. World Drug Report 4. 3 Pereira , J. , Lawlor , P. , Vigano , A. , Dorgan , M. , Bruera , E. ( 2001 ) Equianalgesic dose ratios for opioids. a critical review and proposals for long-term dosing . Journal of Pain and Symptom Management , 22 , 672 – 687 . Google Scholar Crossref Search ADS PubMed 4 EMCDDA . ( 2015 ). European Drug Report - Trends and Developments. 5 Skulska , A. , Kala , M. , Parczewski , A. ( 2005 ) Fentanyl and its analogues in clinical and forensic toxicology . Przeglad Lekarski , 62 , 581 – 584 . Google Scholar PubMed 6 Levine , B. Principles of Forensic Toxicology . Washington, DC: AACC Press , 2009 . 7 Tiscione , N.B. , Wegner , K. ( 2017 ) Validation of the Neogen(R) fentanyl ELISA kit for blood and urine . Journal of Analytical Toxicology , 41 , 313 – 317 . Google Scholar Crossref Search ADS PubMed 8 EMEA . ( 2015 ). Guideline on bioanalytical method validation. EMEA/CHMP/EWP/192217/2009 Rev.1 Corr.2**. 9 SOFT/AAFS . ( 2006 ). Forensic Laboratory Guidelines. Accreditation and Quality Assurance. 10 Schwope , D.M. , Milman , G. , Huestis , M.A. ( 2010 ) Validation of an enzyme immunoassay for detection and semiquantification of cannabinoids in oral fluid . Clinical Chemistry , 56 , 1007 – 1014 . Google Scholar Crossref Search ADS PubMed 11 Guerrieri , D. , Rapp , E. , Roman , M. , Druid , H. , Kronstrand , R. ( 2017 ) Postmortem and toxicological findings in a series of furanylfentanyl-related deaths . Journal of Analytical Toxicology , 41 , 242 – 249 . Google Scholar PubMed 12 Guerrieri , D. , Rapp , E. , Roman , M. , Thelander , G. , Kronstrand , R. ( 2017 ) Acrylfentanyl: another new psychoactive drug with fatal consequences . Forensic Science International , 277 , e21 – e29 . Google Scholar Crossref Search ADS PubMed 13 US Pharmacopoeial Convention . United States Pharmacopeial Convention . Rockville, MD: USP Convention , 2012 . 14 Vardanyan , R.S. , Hruby , V.J. ( 2014 ) Fentanyl-related compounds and derivatives: current status and future prospects for pharmaceutical applications . Future Medicinal Chemistry , 6 , 385 – 412 . Google Scholar Crossref Search ADS PubMed 15 Weltrowska , G. , Chung , N.N. , Lemieux , C. , Guo , J. , Lu , Y. , Wilkes , B.C. , et al. . ( 2010 ) “Carba”-analogues of fentanyl are opioid receptor agonists . Journal of Medicinal Chemistry , 53 , 2875 – 2881 . Google Scholar Crossref Search ADS PubMed 16 Feinberg , A.P. , Creese , I. , Snyder , S.H. ( 1976 ) The opiate receptor: a model explaining structure-activity relationships of opiate agonists and antagonists . Proceedings of the National Academy of Sciences of the United States of America , 73 , 4215 – 4219 . Google Scholar Crossref Search ADS PubMed 17 Petrov , R.R. , Vardanyan , R.S. , Lee , Y.S. , Ma , S.W. , Davis , P. , Begay , L.J. , et al. . ( 2006 ) Synthesis and evaluation of 3-aminopropionyl substituted fentanyl analogues for opioid activity . Bioorganic & Medicinal Chemistry Letters , 16 , 4946 – 4950 . Google Scholar Crossref Search ADS PubMed 18 Van Daele , P.G. , De Bruyn , M.F. , Boey , J.M. , Sanczuk , S. , Agten , J.T. , Janssen , P.A. ( 1976 ) Synthetic analgesics: N-(1-[2-arylethyl]-4-substituted 4-piperidinyl) N-arylalkanamides . Arzneimittel-Forschung , 26 , 1521 – 1531 . Google Scholar PubMed 19 McIntyre , I.M. , Trochta , A. , Gary , R.D. , Malamatos , M. , Lucas , J.R. ( 2015 ) An acute acetyl fentanyl fatality: a case report with postmortem concentrations . Journal of Analytical Toxicology , 39 , 490 – 494 . Google Scholar Crossref Search ADS PubMed 20 Fort , C. , Curtis , B. , Nichols , C. , Niblo , C. ( 2016 ) Acetyl fentanyl toxicity: two case reports . Journal of Analytical Toxicology , 40 , 754 – 757 . Google Scholar PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Mazumder, Shrila; Ahamed, Rayaj A; McGahee, Ernest; Wang, Lanqing; Seyler, Tiffany H

doi: 10.1093/jat/bky045pmid: 30010885

Abstract Cigarette smoking significantly increases the risk of cancer and cardiovascular diseases as well as premature death. Aromatic amines (AAs) such as o-toluidine, 2-aminonaphthalene and 4-aminobiphenyl are found in cigarette smoke and are well-established human bladder carcinogens presumably acting via the formation of DNA adducts. These amines may be metabolized in the liver to acetylated or glucuronidated forms or oxidized to a hydroxylamine which may react with protein and DNA to form adducts. Free, acetylated and glucuronidated AAs are excreted in urine and can be measured as exposure biomarkers. Using isotope dilution GC–MS/MS, our laboratory quantifies six urinary AAs that are known or suspected carcinogens—o-toluidine, 2,6-dimethylaniline, o-anisidine, 1-aminonaphthalene, 2-aminonaphthalene and 4-aminobiphenyl—for large population studies such as the National Health and Nutrition Examination Survey (NHANES). We also monitor two additional corresponding structural isomers—2-aminobiphenyl and 3-aminobiphenyl—to verify isomer separation. A new and improved automated sample preparation method was developed to quantify these AAs, in which, sample cleanup was done via Supported Liquid Extraction (SLE+ ISOLUTE®) on a Hamilton STAR™ workstation. This automated method increased sample throughput by reducing sample cleanup time from 8 to 4 h while maintaining precision (intra and inter-run coefficient of variation <7%) and accuracy (±17%). Recent improvements in our GC/MS method have enhanced our assay sensitivity and specificity, resulting in longer analytical column life and maintaining or reducing the limit of detection for all six analytes. Indigo ASCENTTM software (3.7.1, Indigo BioAutomation, Inc.) is used for peak integration, calibration and quantification. A streamlined sample data flow was created in parallel with the automated method, in which samples can be tracked from receiving to final laboratory information management system output with minimal human intervention, minimizing potential human error. This newly validated, automated method and sample data flow are currently applied in biomonitoring of AAs in the US noninstitutionalized population NHANES 2013–2014 cycle. aromatic amines, automation, sample data flow, gas chromatography, tandem mass spectrometry Introduction Cigarette smoking is a significant risk factor for cancer for both smokers and nonsmokers (1). Exposure of nonsmokers to secondhand tobacco smoke (SHS) has been linked to an increased risk of cancer, coronary heart disease and respiratory illnesses in both adults and children (2–8). As SHS causes premature death and disease in nonsmokers, its presence in the environment remains a significant public health concern (9). SHS is mainly composed of gases and particulate matters generated from a mixture of sidestream smoke, which is emitted from smoldering cigarettes between puffs and from exhaled mainstream smoke. The Surgeon General concluded in 2006 that there is no risk-free level of exposure to SHS. Smoking tobacco and inhaling SHS may be major sources of exposure to several aromatic amines (AAs) (10–12), which are suggested to be principal agents for the development of bladder cancer in humans (13). The total nitrogen content in tobacco leaves—as derived from nitrates, ammonia, amino acids, amides and alkaloids—ultimately contributes to the formation of AAs in tobacco smoke (14–16). Nitrate, which is introduced to the growing tobacco plant through the application of fertilizer, can be converted to ammonia, which, in turn, is converted to other nitrogenous organic compounds such as amino acids. Intermediate NH2 radicals, forming during the pyrolysis of ammonia during tobacco combustion, may react with aromatic CH groups (from compounds already present in the tobacco leaves) to form the AAs (17). In addition to the pyrosynthetic mechanism, AAs may also be transferred directly from the tobacco leaves into the smoke via thermal degradation of alkaloids and amino acids (18, 19). In addition to tobacco smoke, other exposure sources of AAs include several chemical industry sectors such as dyes and pigments (e.g., azo dyes, indigo dyes), pharmaceuticals, pesticides, herbicides, synthetic rubber and plastics (20–24). In manufacturing these industrial chemicals, AAs are used as raw materials or intermediates, and therefore, they should not occur in the final products. As such, occupational exposure to AAs can occur by inhalation or skin contact during the production of chemicals that use AAs as raw materials or intermediates. AAs can also be found in environmental pollution such as diesel exhaust, combustion of wood chips and rubber and substances in charcoal barbequed meats and fish (25, 26). Natural occurence of AA was reported, for example, in the aroma components of black tea and certain vegetables (27, 28). Other potential sources include emissions from cooking oils (e.g., vegetable, sunflower and refined lard oil) (29). The International Agency for Research on Cancer has classified several AAs as carcinogenic to humans (Group 1) or possibly carcinogenic to humans (Group 2B): AAs such as o-toluidine (OTOL), 2-aminonapthalene and 4-aminobiphenyl, for example, are well-established human bladder carcinogens (13, 30, 31). Several AAs are on the FDA’s list of harmful and potentially harmful constituents since the passage of The Family Smoking Prevention and Tobacco Control Act in 2009 and the creation of the Center of Tobacco Products (32). During smoke inhalation, AAs are first carried into the bloodstream from the lungs, metabolized by the liver, then travels to the bladder to be eventually excreted out of the body through urination. The amine functional groups may be metabolized in the liver to the acetylated and/or glucuronidated forms, or they may be oxidized to a hydroxylamine form, which undergoes further conversion via an acetylation reaction to form an N-acetoxy metabolite. The N-acetoxy metabolite is able to undergo nonenzymatic breakdown to yield the reactive nitrenium ion, nitrene or a free radical that can covalently bind to tissue macromolecules (proteins) and DNA to form adducts (20, 33, 34). Formation of the hydroxylamine metabolite is considered the primary pathway to AA carcinogenicity in the bladder (20), whereas the acetylated and glucuronidated forms of AAs (excreted in urine) are the products of the body’s detoxification metabolism pathway. As such, the free, acetylated and glucuronidated AAs excreted in urine allow them to be surrogate biomarkers of AA exposure. These compounds have been traditionally measured by manual approaches involving laborious and time-consuming sample preparation steps. Intricate sample cleanup—often utilizing hydrolysis, direct liquid–liquid extraction and/or a solid-phase extraction step—is necessary due to the complexity of the matrix analyzed (e.g., smoke, wastewater, urine, serum, breast milk) and the ultralow levels of AAs (4, 5, 14, 35–43). Acid hydrolysis—with hydrochloric or sulfuric acid—followed by basification of hydrolysate, is the most common step taken toward deconjugating AAs in urine samples; methods involving enzyme and base hydrolysis have also been validated (4, 43). As for sample analysis, gas chromatography (GC)—following a derivatization pretreatment step—is the most common separation technique, though some labs have developed various liquid chromatography (LC) methods (4, 38, 40, 41). Detection systems range from single-, triple-quadrupole or orbitrap mass analyzers to flame ionization detector (35), UV–Vis spectrophotometer (41), electron-capture detector (5) and electrochemical detector (4), where mass spectrometry is the most commonly used detection technique (36–40, 42, 43). Electron impact (EI) ionization, negative-mode chemical ionization applying methane as reactant gas (33, 36) and positive-mode electrospray ionization (ESI+) (38) are all common ionization methods applied in conjunction with the various GC-MS or LC-MS methods developed to detect AAs. In this paper, we present a newly validated method that replaces our previous manual sample preparation method (44) with an automated approach using a Hamilton STAR™ workstation. It is the first time an automated sample preparation approach is reported. In addition, we also report a new streamlined sample data flow created in parallel to the automated sample preparation method, in which samples can be tracked from receiving to final laboratory information management system (LIMS) output with minimal human intervention. In our new method, six AAs are quantified: o-toluidine, 2,6-dimethylaniline (26DM), o-anisidine (OANS), 1-aminonaphthalene (1AMN), 2-aminonaphthalene (2AMN) and 4-aminobiphenyl (4ABP). Two additional related structural isomers are monitored to ensure isomeric separation: 2-aminobiphenyl (2ABP) and 3-aminobiphenyl (3ABP) (Figure 1). Figure 1. Open in new tabDownload slide Chemical structures of eight AAs. Analytes that are quantified are indicated with an asterisk. Method and materials Materials Native (unlabeled) standards used to make calibration curves were purchased from Fluka, Aldrich and Sigma (Sigma-Aldrich, St. Louis, MO). OTOL-13C6, OANS-2H7, 1AMN-2H9 were purchased from Medical Isotope (Pelhem, NH); 2AMN-2H7 was purchased from CDN Isotopes (Pointe-Claire, Quebec, Canada); 4ABP-2H9 was purchased from Cambridge Isotope Laboratory (Andover, MA) and 26DM-2H6 was purchased from Toronto Research Chemical (North York, Canada). All native standards used for the second source accuracy test were purchased from Toronto Research Chemical, except for 4ABP, which was purchased from Sigma. Sodium hydroxide pellets (semiconductor grade) were purchased from Sigma. High-purity hydrochloric acid, pentafluoropropionic anhydride (PFPA) and trimethylamine hydrochloride (TMA-HCl) were obtained from Sigma-Aldrich. All solvents were GC2 grade, except for water, which was HPLC-grade. All solvents were purchased from Burdick and Jackson Labs (distributed by VWR, Suwanee, GA), and all gases were ultrahigh purity grade. Isolute™ support liquid extraction (SLE) cartridges were purchased from Biotage (Charlotte, NC). Vials of 4.5-mL high recovery samples were purchased from ChemGlass (Vineland, NJ). EP Scientific 10- mL silanized glass tubes were purchased from LabDepot (Dawsonville, GA). Wheaton 1-mL amber crimp GC vials with 300-μL insert and SUN-Sri 11-mm aluminum crimp caps with rubber septum were ordered from ThermoFisher Scientific (Suwanee, GA). All GC–MS/MS supplies were purchased from Agilent Technologies (Santa Clara, CA); all filter tips for sample aliquoting in the Hamilton Microlab STAR™ Liquid Handling Workstation were purchased from the Hamilton Company (Reno, NV). Instrument All sample and internal standard aliquoting was performed on the Hamilton Microlab STAR™ Liquid Handling Workstation, which was customized with a recessed deck to handle the physical dimensions of high-volume cartridges (Figure 2). The Hamilton STAR was configured with four 5-mL channels and eight 1-mL channels, compression-induced O-Ring expansion (CO-RE) paddle grippers, autoload and barcode reader and four custom-built deep well vacuum chambers. The 5-mL pipetting channels used 4.5-mL CO-RE tips with filters and the eight pipetting channels utilized both 1000- and 50-μL CO-RE tips with filters. Figure 2. Open in new tabDownload slide Schematic for Hamilton STAR with a customized receding deck that can accommodate high cartridge sample preparation: (a) custom 10-mL cartridge holders; (b) custom 10-mL tube holders; (c) eight 1-mL channels; (d) four 5-mL channels; (e) CO-RE paddle grippers; (f) barcode reader; (g) custom vacuum chambers. Automated sample preparation Urine samples were transferred from cryovials in 2 mL aliquots, along with approximately 450 pg of internal standard (45 μL of 10 pg/μL internal standard solution in ethanol) into high recovery vials by the Hamilton STAR liquid handling system. All samples were aspirated using capacitance liquid-level detection and dispensed using jet empty settings. Samples were mixed (consisting of a rapid aspirate and dispense cycle of 500 μL into the same container) thrice prior to being transferred to the high recovery vials to ensure uniformity. The Hamilton Autoload and Barcode Scanner were used to scan and decode Code 128 barcodes affixed to the original sample containers. A Microsoft Excel (Microsoft Corporation, Redmond, WA) output file of the scanned vials was generated to assure sample tracking and placement. The delivery volumes of 45 μL and 2 mL were verified gravimetrically with less than 1% error using the Hamilton Volume Verification Kit (Reno, NV) and a Mettler-Toledo High Precision Weight Module, Model WXS205SDU (Columbus, OH). Urine samples were hydrolyzed with 50 μL of 10 M NaOH and incubated at 90°C for approximately 15 h on a hot plate (VWR). After cooling to room temperature, the samples were relocated to the Hamilton deck, where the total volume was transferred onto Isolute™ SLE cartridges. These cartridges were placed in custom 10-mL cartridge holders over the deep well vacuum chambers. Silanized glass tubes, placed in custom tube holders, were positioned under the Isolute SLE cartridges using the CO-RE paddle grippers. Analytes were then eluted with three washes of 3 mL dichloromethane and collected into the silanized glass tubes. The silanized glass tubes were then transferred to a Thermo Scientific Savant SPD2010 SpeedVac Concentrator (Holbrook, NY), where the contents were evaporated to approximately 250 μL. To the concentrated eluate, 3 μL of 1.0 M TMA (1.0 g of 98%-purity TMA-HCl dissolved in 2 mL water, neutralized with 1–3 μL of 10 M NaOH, then extracted TMA with 5 mL hexane) and 3 μL of PFPA were added and kept capped at room temperature for 30 min to complete derivatization of the AAs. Afterward, the derivatized samples were manually transferred via pipette to 1-mL amber GC vials with 300 μL insert and further evaporated to completion in the Savant. An amount of 10 μL of toluene was added to each vial to reconstitute the sample. The vials were subsequently capped and vortexed before being stored in −20°C or analyzed on the GC–MS/MS. GC–MS/MS analysis All analyses were performed on two Agilent 7890-7000 C GC–MS/MS. The 7890 GCs were equipped with multimode inlet and Agilent single taper liner (4 mm ID) with glass wool. The use of liners with glass wool helped minimize contamination of the column with sample residues while maintaining the signal sensitivities of all analytes (as compared to the signal sensitivities recorded from injections made on a single taper plain liner). The injection mode was pulsed splitless at 30 psi for 0.35 min to minimize diffusion and improve analyte transfer onto the analytical column. The injection volume was 1 μL, and the injection port temperature was held at 250°C for the duration of each analytical run. A two-column setup connected by a purged union was used: the analytical column used was the Agilent J&W DB-FFAP (30 m × 0.25 mm × 0.25 μm), which is composed of a nitroterephthalic-acid-modified polyethylene glycol stationary phase of high polarity; the second column in series (“postcolumn”) was the Agilent inert fused silica (1.0 m × 0.15 mm), which connected the analytical column to the MS transfer line. The GC oven was initially set to 80°C for 2 min after sample injection, then heated to 180°C at 30°C/min ramp rate, then to the final temperature of 240°C at 15°C/min ramp rate. Each analytical run was operated under a constant 1.68 mL/min flow rate on the analytical column and 1.85 mL/min on the postcolumn, using helium as the carrier gas. To remove high-boiling sample matrix components from the analytical column, reduce interferences in subsequent sample injections and increase the overall column life, a postcolumn backflushing method was included. The backflush was setup for five void volumes, keeping a −2.20 mL/min constant back flow on the analytical column at 240°C. The transfer line and MS source temperatures were both at 280°C. The MS source mode was positive EI ionization. EI mass spectra were obtained at ionization energy of 70 eV and gain was set to 20 (×105). For quantitative analysis of the AAs, multiple reaction monitoring was chosen. Ultrahigh purity nitrogen was used as the collision gas, and ultrahigh purity helium was used as the quenching gas. Details of instrument operational conditions and the ions used for quantitation are listed in Supplementary Tables S1 and S2. Sample data flow An automated system for tracking sample data was created in parallel with the automated sample preparation method (Figure 3). Samples are received and logged into the LIMS reporting system before they are queued for preparation and analysis. Samples to be prepared were scanned by the Hamilton STAR, which, upon completion, generated an output file. The Hamilton output file was run through an Excel macro which generated two modified sequence files: one to be imported into Agilent MassHunter and one to be uploaded to Indigo ASCENT™ once Agilent GC/MS analyses were completed. Once the raw data were acquired from the GC–MS/MS, they could either be analyzed in Agilent MassHunter Quantitative software or uploaded to Indigo ASCENT™ for automatic peak integration. A formatted output file containing the final calculated concentration data was generated and was directly uploaded to the LIMS system. Unknown samples were evaluated individually according to set quality assurance (QA) rules, including difference in retention times of internal standard and native ion transition peaks, confirmation ion ratio, internal standard peak area and concentration exceeding calibration dynamic range. Individual samples and/or analyte would be flagged for repeat or dilution if any of the QA rules were violated. Batch quality controls (QCs) were evaluated according to modified Westgard QC rules (45). A batch would be rejected and repeated if any of the QC rules were violated. In addition, the blank was examined, and the entire batch was rejected and repeated if blank level exceeded the established limit for each analyte. (The blank limit for each analyte was obtained from 50 individual runs, only one run per day, over a period of several months.) Final results that passed all QA and QC rules were exported to a final reporting system such as National Health and Nutrition Examination Survey (NHANES). Figure 3. Open in new tabDownload slide Sample data flow chart. Results All six quantified AAs (OTOL, 26DM, OANS, 1AMN, 2AMN and 4ABP) and corresponding structural isomers—m-toluidine (MTOL), p-toluidine (PTOL), 2,3-dimethylaniline (23DM), 2,4-dimethylaniline (24DM), 2,5-dimethylaniline (25DM), 3,4-dimethylaniline (34DM), 3,5-dimethylaniline (35DM), m-anisidine (MANS), p-anisidine (PANS), 2ABP, and 3ABP—were separated using GC. Figure 4 shows the total ion counts (TIC) of the six quantified AAs and the 11 structural isomers in a calibration standard, to validate isomeric separation from the target analytes. The TIC of the six quantified AAs and two of the monitored structural isomers (2ABP and 3ABP) in a urine sample is also shown in Figure 4. The high degree of sensitivity and specificity of this method allows detection of all six target analytes at trace levels (parts-per-trillion) in human urine samples. Figure 4. Open in new tabDownload slide The TIC of six quantified AAs and two of the monitored isomers in an urine sample containing approximately 100 ng/L of each target analyte (represented as dashed line) is overlaid on the TIC of a calibration standard containing 100 pg/μL of each target analyte (represented as solid line). The three insets show the related structural isomers of OANS, OTOL and 26DM, each spiked at 100 pg/μL. Limit of detection Since our previous work (44), in which the limit of detection (LoD) was estimated using the extrapolated, limiting standard deviation obtained from calibration curves, the LoD in this study was determined according to the guideline for determination of LoD by the Clinical and Laboratory Standard Institute (46) using the four LoD pools: LoD0, LoD1, LoD2 and LoD3. The pools were made using filtered nonsmoker urine and spiked at 0, 20, 40 and 60 ng/L for six analytes: OTOL, 26DM, OANS, 1AMN, 2AMN and 4ABP. The analyte LoDs (ng/L) were estimated from 50 independent analytical runs, one analytical run per day. Because OTOL is detected in the blanks, its LoD was calculated using the equation: 3σblank = LoD, where σblank is the standard deviation of calculated blank levels from 54 individual runs. The analyte LoDs are listed in Table I. Table I. Calculated slopes from calibration curves prepared in urine (matrix) and hexane (nonmatrix). Differences in the averages of the three slopes, obtained from each calibration set, are minimal and validates use of a nonmatrix calibration set to quantitate AAs in urine. Analyte . LoD (ng/L) . Hexane and urine calibration range (pg/μL) . Hexane calibration curve, slope . Hexane, averaged slopes . Urine calibration curve, slope . Urine, averaged slopes . Difference in averaged slopes (%) . Run 1 . Run 2 . Run 3 . Run 1 . Run 2 . Run 3 . OTOL 111.2 0.507–1,280 0.01811 0.01673 0.01753 0.01746 0.01768 0.01688 0.01598 0.01685 3.49 26DM 15.7 0.502–100 0.01613 0.01567 0.01563 0.01581 0.01643 0.01573 0.01611 0.01609 −1.76 OANS 7.0 0.504–99.9 0.01164 0.01144 0.01160 0.01156 0.01153 0.01099 0.01141 0.01131 2.19 1AMN 1.5 0.504–99.9 0.03636 0.03736 0.03735 0.03702 0.03553 0.03610 0.03574 0.03579 3.33 2AMN 2.8 0.489–100 0.02788 0.02850 0.02849 0.02829 0.02924 0.02948 0.02924 0.02932 −3.64 4ABP 1.8 0.482–98.6 0.00699 0.00698 0.00673 0.00690 0.00712 0.00716 0.00697 0.00708 −2.62 Analyte . LoD (ng/L) . Hexane and urine calibration range (pg/μL) . Hexane calibration curve, slope . Hexane, averaged slopes . Urine calibration curve, slope . Urine, averaged slopes . Difference in averaged slopes (%) . Run 1 . Run 2 . Run 3 . Run 1 . Run 2 . Run 3 . OTOL 111.2 0.507–1,280 0.01811 0.01673 0.01753 0.01746 0.01768 0.01688 0.01598 0.01685 3.49 26DM 15.7 0.502–100 0.01613 0.01567 0.01563 0.01581 0.01643 0.01573 0.01611 0.01609 −1.76 OANS 7.0 0.504–99.9 0.01164 0.01144 0.01160 0.01156 0.01153 0.01099 0.01141 0.01131 2.19 1AMN 1.5 0.504–99.9 0.03636 0.03736 0.03735 0.03702 0.03553 0.03610 0.03574 0.03579 3.33 2AMN 2.8 0.489–100 0.02788 0.02850 0.02849 0.02829 0.02924 0.02948 0.02924 0.02932 −3.64 4ABP 1.8 0.482–98.6 0.00699 0.00698 0.00673 0.00690 0.00712 0.00716 0.00697 0.00708 −2.62 Open in new tab Table I. Calculated slopes from calibration curves prepared in urine (matrix) and hexane (nonmatrix). Differences in the averages of the three slopes, obtained from each calibration set, are minimal and validates use of a nonmatrix calibration set to quantitate AAs in urine. Analyte . LoD (ng/L) . Hexane and urine calibration range (pg/μL) . Hexane calibration curve, slope . Hexane, averaged slopes . Urine calibration curve, slope . Urine, averaged slopes . Difference in averaged slopes (%) . Run 1 . Run 2 . Run 3 . Run 1 . Run 2 . Run 3 . OTOL 111.2 0.507–1,280 0.01811 0.01673 0.01753 0.01746 0.01768 0.01688 0.01598 0.01685 3.49 26DM 15.7 0.502–100 0.01613 0.01567 0.01563 0.01581 0.01643 0.01573 0.01611 0.01609 −1.76 OANS 7.0 0.504–99.9 0.01164 0.01144 0.01160 0.01156 0.01153 0.01099 0.01141 0.01131 2.19 1AMN 1.5 0.504–99.9 0.03636 0.03736 0.03735 0.03702 0.03553 0.03610 0.03574 0.03579 3.33 2AMN 2.8 0.489–100 0.02788 0.02850 0.02849 0.02829 0.02924 0.02948 0.02924 0.02932 −3.64 4ABP 1.8 0.482–98.6 0.00699 0.00698 0.00673 0.00690 0.00712 0.00716 0.00697 0.00708 −2.62 Analyte . LoD (ng/L) . Hexane and urine calibration range (pg/μL) . Hexane calibration curve, slope . Hexane, averaged slopes . Urine calibration curve, slope . Urine, averaged slopes . Difference in averaged slopes (%) . Run 1 . Run 2 . Run 3 . Run 1 . Run 2 . Run 3 . OTOL 111.2 0.507–1,280 0.01811 0.01673 0.01753 0.01746 0.01768 0.01688 0.01598 0.01685 3.49 26DM 15.7 0.502–100 0.01613 0.01567 0.01563 0.01581 0.01643 0.01573 0.01611 0.01609 −1.76 OANS 7.0 0.504–99.9 0.01164 0.01144 0.01160 0.01156 0.01153 0.01099 0.01141 0.01131 2.19 1AMN 1.5 0.504–99.9 0.03636 0.03736 0.03735 0.03702 0.03553 0.03610 0.03574 0.03579 3.33 2AMN 2.8 0.489–100 0.02788 0.02850 0.02849 0.02829 0.02924 0.02948 0.02924 0.02932 −3.64 4ABP 1.8 0.482–98.6 0.00699 0.00698 0.00673 0.00690 0.00712 0.00716 0.00697 0.00708 −2.62 Open in new tab Precision To determine intrarun and inter-run precision, two pools were used. The pools were prepared in-house from urine collected (with CDC Institutional Review Board approval) from nonsmokers and spiked with a standard solution of native (unlabeled) analytes at approximately 125 and 500 ng/L. The coefficient of variation (CV) for intraday (n = 5) and interday (n = 5) runs are calculated and listed in Table II. Intraday and interday CV values, for all AAs at each spiked concentration, were below 10%. Table III. Short-term (24 h at 23°C), long-term (2 years at −70°C), and thaw and re-freeze (T/RF) stability of unprocessed samples; and stability of processed samples at 10°C (72 h) and −20°C (13 weeks). Analyte . Nominal concentration (ng/L) . Measured concentration (ng/L) . Unprocessed samples at −70°C for 2 year (%−diff) . Unprocessed samples at 23°C for 24 h (%−diff) . T/RF stability of unprocessed samples . −20°C storage stability of processed samples . 10°C storage stability of processed samples . Control . Five T/RF cycles (%−diff) . Day 1 . Week 1 (%−diff) . Week 13 (%−diff) . Day 1 . Autosampler for 24 h (%−diff) . Autosampler for 72 h (%−diff) . OTOL 150.7 168.9 (12.10) 218.0 198.2 (−9.09) 57.0 60.7 (6.55) 53.2 (−6.72) 623.0 588.5 (−5.53) 562.8 (−9.66) 485.9 493.0 (1.48) 511.3 (5.24) 67.6 59.7 (−11.61) 63.6 (−5.86) 82.2 87.4 (6.38) 92.3 (12.25) 168.0 168.7 (0.40) 181.1 (7.79) 490.2 498.9 (1.75) 501.3 (2.26) 26DM 121.3 109.8 (−9.46) 157.5 148.9 (−5.46) 19.8 17.8 (−10.28) 17.6 (−10.92) 459.6 426.9 (−7.11) 418.3 (−8.98) 502.6 492.6 (−2.01) 519.1 (3.27) 43.8 43.0 (−1.95) 38.0 (−13.38) 68.1 61.4 (−9.89) 59.6 (−12.45) 149.6 147.3 (−1.57) 150.6 (0.63) 543.4 480.5 (−11.58) 499.1 (−8.16) OANS 153.4 148.0 (−3.48) 158.2 144.4 (−8.71) 44.9 46.4 (3.25) 54.0 (20.39) 427.1 420.2 (−1.60) 405.1 (−5.15) 505.8 498.5 (−1.44) 488.8 (−3.37) 70.5 66.4 (−5.88) 64.1 (−9.05) 92.8 83.0 (−10.51) 86.4 (−6.87) 153.5 137.6 (−10.39) 141.5 (−7.81) 544.4 531.9 (−2.28) 516.4 (−5.14) 1AMN 110.8 120.0 (8.31) 109.5 91.5 (−16.39) 13.6 15.5 (14.37) 14.1 (3.62) 269.2 287.7 (6.87) 303.4 (12.71) 458.1 496.4 (8.36) 434.9 (−5.08) 34.1 33.7 (−1.30) 31.8 (−6.84) 43.8 53.1 (21.19) 50.7 (15.84) 94.1 115.7 (22.87) 114.2 (21.25) 449.4 472.3 (5.08) 501.7 (11.64) 2AMN 111.7 116.7 (4.45) 101.4 90.3 (−10.93) 14.3 15.1 (5.54) 15.2 (6.70) 318.8 321.3 (0.81) 323.7 (1.54) 479.8 511.5 (6.62) 476.7 (−0.63) 35.4 38.1 (7.76) 37.7 (6.46) 53.4 56.8 (6.35) 56.7 (6.27) 96.1 100.7 (4.72) 99.4 (3.42) 528.8 513.3 (−2.93) 509.1 (−3.71) 4ABP 109.0 111.9 (2.64) 111.6 105.2 (−5.75) 21.5 20.3 (−5.36) 21.5 (0.00) 347.0 339.9 (−2.06) 342.0 (−1.46) 446.1 467.5 (4.80) 428.2 (−3.99) 42.5 41.3 (−2.65) 42.1 (−0.90) 62.4 58.4 (−6.43) 61.1 (−2.17) 120.2 106.0 (−11.81) 108.1 (−10.04) 476.0 461.0 (−3.15) 476.0 (0.00) Analyte . Nominal concentration (ng/L) . Measured concentration (ng/L) . Unprocessed samples at −70°C for 2 year (%−diff) . Unprocessed samples at 23°C for 24 h (%−diff) . T/RF stability of unprocessed samples . −20°C storage stability of processed samples . 10°C storage stability of processed samples . Control . Five T/RF cycles (%−diff) . Day 1 . Week 1 (%−diff) . Week 13 (%−diff) . Day 1 . Autosampler for 24 h (%−diff) . Autosampler for 72 h (%−diff) . OTOL 150.7 168.9 (12.10) 218.0 198.2 (−9.09) 57.0 60.7 (6.55) 53.2 (−6.72) 623.0 588.5 (−5.53) 562.8 (−9.66) 485.9 493.0 (1.48) 511.3 (5.24) 67.6 59.7 (−11.61) 63.6 (−5.86) 82.2 87.4 (6.38) 92.3 (12.25) 168.0 168.7 (0.40) 181.1 (7.79) 490.2 498.9 (1.75) 501.3 (2.26) 26DM 121.3 109.8 (−9.46) 157.5 148.9 (−5.46) 19.8 17.8 (−10.28) 17.6 (−10.92) 459.6 426.9 (−7.11) 418.3 (−8.98) 502.6 492.6 (−2.01) 519.1 (3.27) 43.8 43.0 (−1.95) 38.0 (−13.38) 68.1 61.4 (−9.89) 59.6 (−12.45) 149.6 147.3 (−1.57) 150.6 (0.63) 543.4 480.5 (−11.58) 499.1 (−8.16) OANS 153.4 148.0 (−3.48) 158.2 144.4 (−8.71) 44.9 46.4 (3.25) 54.0 (20.39) 427.1 420.2 (−1.60) 405.1 (−5.15) 505.8 498.5 (−1.44) 488.8 (−3.37) 70.5 66.4 (−5.88) 64.1 (−9.05) 92.8 83.0 (−10.51) 86.4 (−6.87) 153.5 137.6 (−10.39) 141.5 (−7.81) 544.4 531.9 (−2.28) 516.4 (−5.14) 1AMN 110.8 120.0 (8.31) 109.5 91.5 (−16.39) 13.6 15.5 (14.37) 14.1 (3.62) 269.2 287.7 (6.87) 303.4 (12.71) 458.1 496.4 (8.36) 434.9 (−5.08) 34.1 33.7 (−1.30) 31.8 (−6.84) 43.8 53.1 (21.19) 50.7 (15.84) 94.1 115.7 (22.87) 114.2 (21.25) 449.4 472.3 (5.08) 501.7 (11.64) 2AMN 111.7 116.7 (4.45) 101.4 90.3 (−10.93) 14.3 15.1 (5.54) 15.2 (6.70) 318.8 321.3 (0.81) 323.7 (1.54) 479.8 511.5 (6.62) 476.7 (−0.63) 35.4 38.1 (7.76) 37.7 (6.46) 53.4 56.8 (6.35) 56.7 (6.27) 96.1 100.7 (4.72) 99.4 (3.42) 528.8 513.3 (−2.93) 509.1 (−3.71) 4ABP 109.0 111.9 (2.64) 111.6 105.2 (−5.75) 21.5 20.3 (−5.36) 21.5 (0.00) 347.0 339.9 (−2.06) 342.0 (−1.46) 446.1 467.5 (4.80) 428.2 (−3.99) 42.5 41.3 (−2.65) 42.1 (−0.90) 62.4 58.4 (−6.43) 61.1 (−2.17) 120.2 106.0 (−11.81) 108.1 (−10.04) 476.0 461.0 (−3.15) 476.0 (0.00) Open in new tab Table III. Short-term (24 h at 23°C), long-term (2 years at −70°C), and thaw and re-freeze (T/RF) stability of unprocessed samples; and stability of processed samples at 10°C (72 h) and −20°C (13 weeks). Analyte . Nominal concentration (ng/L) . Measured concentration (ng/L) . Unprocessed samples at −70°C for 2 year (%−diff) . Unprocessed samples at 23°C for 24 h (%−diff) . T/RF stability of unprocessed samples . −20°C storage stability of processed samples . 10°C storage stability of processed samples . Control . Five T/RF cycles (%−diff) . Day 1 . Week 1 (%−diff) . Week 13 (%−diff) . Day 1 . Autosampler for 24 h (%−diff) . Autosampler for 72 h (%−diff) . OTOL 150.7 168.9 (12.10) 218.0 198.2 (−9.09) 57.0 60.7 (6.55) 53.2 (−6.72) 623.0 588.5 (−5.53) 562.8 (−9.66) 485.9 493.0 (1.48) 511.3 (5.24) 67.6 59.7 (−11.61) 63.6 (−5.86) 82.2 87.4 (6.38) 92.3 (12.25) 168.0 168.7 (0.40) 181.1 (7.79) 490.2 498.9 (1.75) 501.3 (2.26) 26DM 121.3 109.8 (−9.46) 157.5 148.9 (−5.46) 19.8 17.8 (−10.28) 17.6 (−10.92) 459.6 426.9 (−7.11) 418.3 (−8.98) 502.6 492.6 (−2.01) 519.1 (3.27) 43.8 43.0 (−1.95) 38.0 (−13.38) 68.1 61.4 (−9.89) 59.6 (−12.45) 149.6 147.3 (−1.57) 150.6 (0.63) 543.4 480.5 (−11.58) 499.1 (−8.16) OANS 153.4 148.0 (−3.48) 158.2 144.4 (−8.71) 44.9 46.4 (3.25) 54.0 (20.39) 427.1 420.2 (−1.60) 405.1 (−5.15) 505.8 498.5 (−1.44) 488.8 (−3.37) 70.5 66.4 (−5.88) 64.1 (−9.05) 92.8 83.0 (−10.51) 86.4 (−6.87) 153.5 137.6 (−10.39) 141.5 (−7.81) 544.4 531.9 (−2.28) 516.4 (−5.14) 1AMN 110.8 120.0 (8.31) 109.5 91.5 (−16.39) 13.6 15.5 (14.37) 14.1 (3.62) 269.2 287.7 (6.87) 303.4 (12.71) 458.1 496.4 (8.36) 434.9 (−5.08) 34.1 33.7 (−1.30) 31.8 (−6.84) 43.8 53.1 (21.19) 50.7 (15.84) 94.1 115.7 (22.87) 114.2 (21.25) 449.4 472.3 (5.08) 501.7 (11.64) 2AMN 111.7 116.7 (4.45) 101.4 90.3 (−10.93) 14.3 15.1 (5.54) 15.2 (6.70) 318.8 321.3 (0.81) 323.7 (1.54) 479.8 511.5 (6.62) 476.7 (−0.63) 35.4 38.1 (7.76) 37.7 (6.46) 53.4 56.8 (6.35) 56.7 (6.27) 96.1 100.7 (4.72) 99.4 (3.42) 528.8 513.3 (−2.93) 509.1 (−3.71) 4ABP 109.0 111.9 (2.64) 111.6 105.2 (−5.75) 21.5 20.3 (−5.36) 21.5 (0.00) 347.0 339.9 (−2.06) 342.0 (−1.46) 446.1 467.5 (4.80) 428.2 (−3.99) 42.5 41.3 (−2.65) 42.1 (−0.90) 62.4 58.4 (−6.43) 61.1 (−2.17) 120.2 106.0 (−11.81) 108.1 (−10.04) 476.0 461.0 (−3.15) 476.0 (0.00) Analyte . Nominal concentration (ng/L) . Measured concentration (ng/L) . Unprocessed samples at −70°C for 2 year (%−diff) . Unprocessed samples at 23°C for 24 h (%−diff) . T/RF stability of unprocessed samples . −20°C storage stability of processed samples . 10°C storage stability of processed samples . Control . Five T/RF cycles (%−diff) . Day 1 . Week 1 (%−diff) . Week 13 (%−diff) . Day 1 . Autosampler for 24 h (%−diff) . Autosampler for 72 h (%−diff) . OTOL 150.7 168.9 (12.10) 218.0 198.2 (−9.09) 57.0 60.7 (6.55) 53.2 (−6.72) 623.0 588.5 (−5.53) 562.8 (−9.66) 485.9 493.0 (1.48) 511.3 (5.24) 67.6 59.7 (−11.61) 63.6 (−5.86) 82.2 87.4 (6.38) 92.3 (12.25) 168.0 168.7 (0.40) 181.1 (7.79) 490.2 498.9 (1.75) 501.3 (2.26) 26DM 121.3 109.8 (−9.46) 157.5 148.9 (−5.46) 19.8 17.8 (−10.28) 17.6 (−10.92) 459.6 426.9 (−7.11) 418.3 (−8.98) 502.6 492.6 (−2.01) 519.1 (3.27) 43.8 43.0 (−1.95) 38.0 (−13.38) 68.1 61.4 (−9.89) 59.6 (−12.45) 149.6 147.3 (−1.57) 150.6 (0.63) 543.4 480.5 (−11.58) 499.1 (−8.16) OANS 153.4 148.0 (−3.48) 158.2 144.4 (−8.71) 44.9 46.4 (3.25) 54.0 (20.39) 427.1 420.2 (−1.60) 405.1 (−5.15) 505.8 498.5 (−1.44) 488.8 (−3.37) 70.5 66.4 (−5.88) 64.1 (−9.05) 92.8 83.0 (−10.51) 86.4 (−6.87) 153.5 137.6 (−10.39) 141.5 (−7.81) 544.4 531.9 (−2.28) 516.4 (−5.14) 1AMN 110.8 120.0 (8.31) 109.5 91.5 (−16.39) 13.6 15.5 (14.37) 14.1 (3.62) 269.2 287.7 (6.87) 303.4 (12.71) 458.1 496.4 (8.36) 434.9 (−5.08) 34.1 33.7 (−1.30) 31.8 (−6.84) 43.8 53.1 (21.19) 50.7 (15.84) 94.1 115.7 (22.87) 114.2 (21.25) 449.4 472.3 (5.08) 501.7 (11.64) 2AMN 111.7 116.7 (4.45) 101.4 90.3 (−10.93) 14.3 15.1 (5.54) 15.2 (6.70) 318.8 321.3 (0.81) 323.7 (1.54) 479.8 511.5 (6.62) 476.7 (−0.63) 35.4 38.1 (7.76) 37.7 (6.46) 53.4 56.8 (6.35) 56.7 (6.27) 96.1 100.7 (4.72) 99.4 (3.42) 528.8 513.3 (−2.93) 509.1 (−3.71) 4ABP 109.0 111.9 (2.64) 111.6 105.2 (−5.75) 21.5 20.3 (−5.36) 21.5 (0.00) 347.0 339.9 (−2.06) 342.0 (−1.46) 446.1 467.5 (4.80) 428.2 (−3.99) 42.5 41.3 (−2.65) 42.1 (−0.90) 62.4 58.4 (−6.43) 61.1 (−2.17) 120.2 106.0 (−11.81) 108.1 (−10.04) 476.0 461.0 (−3.15) 476.0 (0.00) Open in new tab Accuracy Accuracy was determined by spiking known amounts of AA standard solution into hexane (accuracy in solution) and urine (accuracy in matrix). The accuracy was calculated using the equation: ((measured [AA] − nominal [AA])/nominal [AA]) × 100%. To test accuracy in solution, five levels of testing calibrators were prepared and run with the hexane calibration curve used for AA quantitation in urine samples. The native AAs used to spike the testing calibrators were purchased from vendors or lot numbers different from ones used for making the hexane calibration curve. As this evaluation is necessary to ensure accuracy of the hexane calibration curve, these testing calibrators are analyzed every time a new set of hexane calibration curve is prepared for analysis, or a new internal standard spiking solution is made. Results are listed in Table II. To test accuracy in matrix, replicates of AA-spiked urine samples, at three different concentrations, were analyzed. All spiked urine samples were prepared as unknown samples and run in triplicate over 2 days. For all samples tested, the calculated accuracy was ±17% for all analytes, and thus, accuracy in matrix tests was acceptable (Table II). Table II. Coefficient of variation (CV), %, of intraday and interday precisions, and %-bias accuracy in hexane solution (nonmatrix) and in urine (matrix). Analyte . Nominal concentration (ng/L) . Intraday precision . . Interday precision . . Nominal concentration (pg/μL) . Accuracy in solution . . Nominal concentration (ng/L) . Accuracy in matrix . . Measured concentration (ng/L), n = 5 . CV . Measured concentration (ng/L), n = 5 . CV . Measured concentration (pg/μL) . %−bias . Measured concentration (ng/L) . %−bias . OTOL 150.7 164.5 5.40 159.8 6.26 2.5 2.7 7.60 116.4 97.6 −16.17 485.9 503.2 5.55 527.3 5.75 7.1 8.7 22.70 223.9 205.4 −8.28 26.2 26.8 2.21 806.0 772.6 −4.14 49.3 50.1 1.62 197.0 211.7 7.48 26DM 121.3 113.9 6.22 115.6 5.88 2.5 2.4 −3.60 55.7 47.8 −14.25 502.6 487.1 3.50 503.0 4.37 7.0 6.9 −1.86 107.1 98.0 −8.49 25.7 25.2 −1.79 385.7 386.1 0.10 40.4 40.8 1.04 161.5 161.3 −0.13 OANS 153.4 149.5 3.61 157.9 2.62 2.5 2.8 13.20 58.5 52.9 −9.47 505.8 506.5 2.78 526.1 3.44 7.1 6.8 −4.79 112.5 99.3 −11.74 25.6 27.5 7.42 404.9 372.9 −7.91 49.7 55.9 12.39 198.7 231.5 16.53 1AMN 110.8 120.0 2.19 118.2 4.48 2.5 2.6 4.69 52.9 45.1 −14.71 458.1 491.9 2.19 492.8 3.34 7.0 7.1 1.17 101.7 91.7 −9.83 26.4 25.7 −2.69 366.1 335.0 −8.50 42.7 47.9 12.21 170.9 176.9 3.55 2AMN 111.7 116.7 1.38 121.0 4.31 2.4 2.4 −1.25 57.8 48.0 −16.86 479.8 505.6 2.44 524.4 3.84 6.8 6.4 −6.47 111.1 100.9 −9.19 25.3 24.7 −2.53 400.0 342.7 −14.33 50.0 47.4 −5.18 200.0 199.8 −0.11 4ABP 109.0 112.7 2.51 116.5 5.12 2.4 2.6 9.17 54.7 47.2 −13.78 446.1 468.3 1.13 479.1 3.31 6.8 6.8 −0.15 105.2 94.7 −10.04 26.0 22.9 −11.85 378.9 353.0 −6.83 50.5 54.8 8.44 202.1 232.1 14.83 Analyte . Nominal concentration (ng/L) . Intraday precision . . Interday precision . . Nominal concentration (pg/μL) . Accuracy in solution . . Nominal concentration (ng/L) . Accuracy in matrix . . Measured concentration (ng/L), n = 5 . CV . Measured concentration (ng/L), n = 5 . CV . Measured concentration (pg/μL) . %−bias . Measured concentration (ng/L) . %−bias . OTOL 150.7 164.5 5.40 159.8 6.26 2.5 2.7 7.60 116.4 97.6 −16.17 485.9 503.2 5.55 527.3 5.75 7.1 8.7 22.70 223.9 205.4 −8.28 26.2 26.8 2.21 806.0 772.6 −4.14 49.3 50.1 1.62 197.0 211.7 7.48 26DM 121.3 113.9 6.22 115.6 5.88 2.5 2.4 −3.60 55.7 47.8 −14.25 502.6 487.1 3.50 503.0 4.37 7.0 6.9 −1.86 107.1 98.0 −8.49 25.7 25.2 −1.79 385.7 386.1 0.10 40.4 40.8 1.04 161.5 161.3 −0.13 OANS 153.4 149.5 3.61 157.9 2.62 2.5 2.8 13.20 58.5 52.9 −9.47 505.8 506.5 2.78 526.1 3.44 7.1 6.8 −4.79 112.5 99.3 −11.74 25.6 27.5 7.42 404.9 372.9 −7.91 49.7 55.9 12.39 198.7 231.5 16.53 1AMN 110.8 120.0 2.19 118.2 4.48 2.5 2.6 4.69 52.9 45.1 −14.71 458.1 491.9 2.19 492.8 3.34 7.0 7.1 1.17 101.7 91.7 −9.83 26.4 25.7 −2.69 366.1 335.0 −8.50 42.7 47.9 12.21 170.9 176.9 3.55 2AMN 111.7 116.7 1.38 121.0 4.31 2.4 2.4 −1.25 57.8 48.0 −16.86 479.8 505.6 2.44 524.4 3.84 6.8 6.4 −6.47 111.1 100.9 −9.19 25.3 24.7 −2.53 400.0 342.7 −14.33 50.0 47.4 −5.18 200.0 199.8 −0.11 4ABP 109.0 112.7 2.51 116.5 5.12 2.4 2.6 9.17 54.7 47.2 −13.78 446.1 468.3 1.13 479.1 3.31 6.8 6.8 −0.15 105.2 94.7 −10.04 26.0 22.9 −11.85 378.9 353.0 −6.83 50.5 54.8 8.44 202.1 232.1 14.83 Open in new tab Table II. Coefficient of variation (CV), %, of intraday and interday precisions, and %-bias accuracy in hexane solution (nonmatrix) and in urine (matrix). Analyte . Nominal concentration (ng/L) . Intraday precision . . Interday precision . . Nominal concentration (pg/μL) . Accuracy in solution . . Nominal concentration (ng/L) . Accuracy in matrix . . Measured concentration (ng/L), n = 5 . CV . Measured concentration (ng/L), n = 5 . CV . Measured concentration (pg/μL) . %−bias . Measured concentration (ng/L) . %−bias . OTOL 150.7 164.5 5.40 159.8 6.26 2.5 2.7 7.60 116.4 97.6 −16.17 485.9 503.2 5.55 527.3 5.75 7.1 8.7 22.70 223.9 205.4 −8.28 26.2 26.8 2.21 806.0 772.6 −4.14 49.3 50.1 1.62 197.0 211.7 7.48 26DM 121.3 113.9 6.22 115.6 5.88 2.5 2.4 −3.60 55.7 47.8 −14.25 502.6 487.1 3.50 503.0 4.37 7.0 6.9 −1.86 107.1 98.0 −8.49 25.7 25.2 −1.79 385.7 386.1 0.10 40.4 40.8 1.04 161.5 161.3 −0.13 OANS 153.4 149.5 3.61 157.9 2.62 2.5 2.8 13.20 58.5 52.9 −9.47 505.8 506.5 2.78 526.1 3.44 7.1 6.8 −4.79 112.5 99.3 −11.74 25.6 27.5 7.42 404.9 372.9 −7.91 49.7 55.9 12.39 198.7 231.5 16.53 1AMN 110.8 120.0 2.19 118.2 4.48 2.5 2.6 4.69 52.9 45.1 −14.71 458.1 491.9 2.19 492.8 3.34 7.0 7.1 1.17 101.7 91.7 −9.83 26.4 25.7 −2.69 366.1 335.0 −8.50 42.7 47.9 12.21 170.9 176.9 3.55 2AMN 111.7 116.7 1.38 121.0 4.31 2.4 2.4 −1.25 57.8 48.0 −16.86 479.8 505.6 2.44 524.4 3.84 6.8 6.4 −6.47 111.1 100.9 −9.19 25.3 24.7 −2.53 400.0 342.7 −14.33 50.0 47.4 −5.18 200.0 199.8 −0.11 4ABP 109.0 112.7 2.51 116.5 5.12 2.4 2.6 9.17 54.7 47.2 −13.78 446.1 468.3 1.13 479.1 3.31 6.8 6.8 −0.15 105.2 94.7 −10.04 26.0 22.9 −11.85 378.9 353.0 −6.83 50.5 54.8 8.44 202.1 232.1 14.83 Analyte . Nominal concentration (ng/L) . Intraday precision . . Interday precision . . Nominal concentration (pg/μL) . Accuracy in solution . . Nominal concentration (ng/L) . Accuracy in matrix . . Measured concentration (ng/L), n = 5 . CV . Measured concentration (ng/L), n = 5 . CV . Measured concentration (pg/μL) . %−bias . Measured concentration (ng/L) . %−bias . OTOL 150.7 164.5 5.40 159.8 6.26 2.5 2.7 7.60 116.4 97.6 −16.17 485.9 503.2 5.55 527.3 5.75 7.1 8.7 22.70 223.9 205.4 −8.28 26.2 26.8 2.21 806.0 772.6 −4.14 49.3 50.1 1.62 197.0 211.7 7.48 26DM 121.3 113.9 6.22 115.6 5.88 2.5 2.4 −3.60 55.7 47.8 −14.25 502.6 487.1 3.50 503.0 4.37 7.0 6.9 −1.86 107.1 98.0 −8.49 25.7 25.2 −1.79 385.7 386.1 0.10 40.4 40.8 1.04 161.5 161.3 −0.13 OANS 153.4 149.5 3.61 157.9 2.62 2.5 2.8 13.20 58.5 52.9 −9.47 505.8 506.5 2.78 526.1 3.44 7.1 6.8 −4.79 112.5 99.3 −11.74 25.6 27.5 7.42 404.9 372.9 −7.91 49.7 55.9 12.39 198.7 231.5 16.53 1AMN 110.8 120.0 2.19 118.2 4.48 2.5 2.6 4.69 52.9 45.1 −14.71 458.1 491.9 2.19 492.8 3.34 7.0 7.1 1.17 101.7 91.7 −9.83 26.4 25.7 −2.69 366.1 335.0 −8.50 42.7 47.9 12.21 170.9 176.9 3.55 2AMN 111.7 116.7 1.38 121.0 4.31 2.4 2.4 −1.25 57.8 48.0 −16.86 479.8 505.6 2.44 524.4 3.84 6.8 6.4 −6.47 111.1 100.9 −9.19 25.3 24.7 −2.53 400.0 342.7 −14.33 50.0 47.4 −5.18 200.0 199.8 −0.11 4ABP 109.0 112.7 2.51 116.5 5.12 2.4 2.6 9.17 54.7 47.2 −13.78 446.1 468.3 1.13 479.1 3.31 6.8 6.8 −0.15 105.2 94.7 −10.04 26.0 22.9 −11.85 378.9 353.0 −6.83 50.5 54.8 8.44 202.1 232.1 14.83 Open in new tab Analytical specificity A high degree of analytical specificity was achieved for each analyte with this assay. Monitoring the correct retention times (±0.001 to 0.03 min relative to corresponding internal standards), confirmation ion ratios (which was obtained daily from the calibration standards and compared against the ratios calculated for each study sample) and precursor/product ion transitions helped ensure a high degree of specificity and minimized the influence from any potential interference(s). Carryover Sample carryover was examined by comparing successive pairs of injections of the highest calibrator (100 or 1,280 pg/μL) or high QC samples, with a toluene solvent blank. (The highest calibrator for OTOL was 1,280 pg/μL to accommodate for higher levels of this analyte detected in unknown samples.) No carryover was observed in the solvent blank after any injection of the highest calibrator or high QC. As a precaution, one toluene solvent blank was injected following injection of a full set of calibration standards and after each QC high sample. A toluene solvent blank was also injected at the beginning of each analytical batch to ensure no system contamination prior to standard and sample analyses. Between each individual injection, the syringe barrel was washed six times by drawing 3 μL of toluene. Recovery Sample matrix effects for each analyte were evaluated. Urine samples were spiked with a known amount of labeled AA internal standards and carried through the sample preparation process, as mentioned above. The percent recovery was calculated as the ratio of the response (peak area counts) of internal standards in urine samples to the responses of internal standard in the calibration standards. The average recovery was greater than 85% for all AAs. Linearity limits We have confirmed linear responses for all analytes (R2 ≥ 0.98) across a broad dynamic range, from 0 to 100 pg/μL (for 26DM, OANS, 1AMN, 2AMN and 4ABP) and from 0 to 1,280 pg/μL (for OTOL), with the lowest nonzero standard concentration being 0.5 pg/μL. The dynamic range for OTOL was extended in order to quantitate higher levels of OTOL in the urine samples. Regressions were calculated by plotting the quotients of the peak areas for each native analyte and that of the labeled analyte (the response ratio) as a function of the nominal concentration; a weighting factor of 1/x was used for all analytes. The calculated concentration of analytes in study samples was reported if the value was within the lowest and highest calibrators. Samples with analyte concentration exceeding the highest calibrator were repeated with appropriate dilution to bring the concentration within the validated dynamic range. Matrix equivalence The influence of urine matrix on the hexane (nonmatrix) calibration curve used for daily quantitation was estimated. A urine calibration curve was prepared by spiking standards of AAs at different levels in blank urine samples. The urine standards were subjected to the same sample preparation protocol as unknown urine samples (as described in the “Automated sample preparation” section above). The calibration curves built in urine matrix were run in parallel with calibration curves prepared in hexane (see Figure 5, for 4ABP). The averages of the slopes (n = 3) of each set of calibration curves were compared to assess the influence of matrix effects on AAs. As shown in Table I, the slope differences for all six analytes were within ±5%. These results indicate that a urine matrix has minimal impact on the quantitation of AAs based on calibration curves prepared in hexane. Figure 5. Open in new tabDownload slide Matrix equivalence of 4ABP in urine (matrix) and hexane (nonmatrix) standards. Hexane curve (n = 1) is plotted with solid circular markers, and urine curve (n = 1) is plotted with solid triangular markers. Calibration curves, from either set of matrix, show good linearity for a broad dynamic range (0.5 to 98 pg/μL). Thaw-refreeze and storage stability The long-term storage stability of analytes in unprocessed samples was tested with spiked urine at low and high concentrations that had been stored at −70°C for 2 years. The results from this study indicate that long-term storage at −70°C has minimal impact on sample integrity (Table III). The effect of repeated thaw and refreeze (T/RF) cycles (−70°C – room temperature) on unprocessed urine samples was determined for each analyte. The results indicate that all six analytes are stable following at least five T/RF cycles, with average sample loss staying within the range of 5–17%. Results for individual analytes are listed in Table III. Processed samples were either immediately analyzed on a GC–MS/MS system or provisionally stored at −20 ± 4°C until they could be analyzed. (During GC–MS/MS analysis, processed samples were stored in a cooled (10°C) autosampler.) To evaluate storage stability at −20°C, five spiked nonsmoker samples were injected after processing. The initial measurements are listed as “Day 1” in Table III. After initial injection, these samples were stored at −20°C for 1 week, then reinjected (“Week 1” measurement) and stored again at −20°C for 13 weeks then reinjected for a final time (“Week 13” measurement). The results from repeated injections of samples stored at −20°C indicate that analytes in processed samples are stable for up to 13 weeks, with a majority of the repeated measurement yielding ±10% difference from initial measurement (Table III). Likewise, the results from repeated injections of a sample stored in the autosampler indicate that analyte levels are not significantly impacted by short-term (24 and 72 h) storage conditions following the initial day (“Day 1”) sample was placed on autosampler and analyzed (Table III). The effect on analyte stability due to the length of time needed to process urine samples at room temperature (23°C ± 1°C) was also assessed. The results listed in Table III show insignificant changes in analyte levels. Ruggedness test To test the ruggedness of both the sample preparation and the GC–MS/MS method, QC samples were prepared and run under varied conditions in five parameters, each tested separately. Ruggedness was tested to determine which of the parameters (if any) would potentially affect assay accuracy. The parameters tested include: (i) injection port temperature; (ii) injection pulse pressure; (iii) amount of PFPA used during derivatization; (iv) sample hydrolysis duration and (v) the lot/work order of DB-FFAP column used for analysis. Samples were tested for all parameters at below, above and normal operating conditions. For all parameters tested, less than 17% difference in concentration was calculated for all analytes, with the majority at or below 10%. The complete set of result is listed in Supplementary Table S3. Instrument cross-validation For our assay, two Agilent GC–MS/MS systems were used to analyze processed samples. For the AA assay, five spiked and one nonspiked nonsmoker urine samples were used for instrument cross-validation. The Pearson coefficient for all six quantified analytes was within the range of 0.97–0.99. QC characterization The two QC pools used for this method—QC low (spiked at approximately 125 ng/L for all analytes) and QC high (spiked at approximately 500 ng/L for all analytes)—were characterized with 39 replicates from each pool, spanning almost 10 months. QC characterization statistics was subsequently used to verify methodological precision for each analytical run according to modified Westgard QC rules (45). No significant changes in concentration were observed for any analyte, in either QC pool, over the duration of the characterization period, or 16 months after the QC pools were prepared (Supplementary Table S4). Validation of Indigo ASCENT™ Indigo ASCENT™ is a software used in our laboratory to automatically integrate chromatograms based on proprietary algorithms. To ensure the accuracy of automatic chromatographic peak integration, quantitated results from 150 samples obtained from Agilent MassHunter Workstation Quantitative Analysis and Indigo ASCENT™ were compared to ensure that the measurements of analyte concentrations were equivalent. A nonparametric test (Mann–Whitney) concluded that the calculated results obtained from Indigo ASCENT™ and MassHunter Workstation were statistically equivalent. Discussion Automation of sample preparation processes became a necessary part of biomonitoring measurements applied to large population studies such as NHANES. Analyzing more than 5,000 samples per 2-year cycle requires increased throughput compared with typical manual sample preparation. The automated workstation allows the analyst to increase the necessary sample throughput by reducing the time it takes to aliquot samples. Compared to our previous manual sample preparation method (44), the new automated sample preparation method effectively reduces the sample cleanup time from 8 to 4 h, while improving and maintaining the high precision of the assay. The LoDs obtained from our method were within the range of 1.5 ng/L (26DM, OANS, 1AMN, 2AMN and 4ABP) to 111.2 ng/L (OTOL), with 1AMN, 2AMN and 4ABP exhibiting detection limits lower than 3 ng/L. With LoDs of most analytes below 16 ng/L, our new method’s detection limits are either lower than (4, 5, 38, 39, 41–43) or comparable to (35–37, 40, 44) the detection limits reported by other groups. It is important to note that the LoDs reported in the literature were estimated based on either the calibration curves used to analyze samples or through analysis of chromatographic peaks that had signal-to-noise ratios greater than 3 (37, 40, 44). For instance, the lowest detection limits for OTOL and 26DM were reported to be 5.2 and 9.3 ng/L, respectively, using aqueous standards to estimate each value (37); the lowest detection limits for 1AMN, 2AMN and 4ABP were reported to be 5.0, 3.0 and 1.5 ng/L, respectively, using solvent standards to estimate each value (40). Analyte LoDs in our assay were obtained from 50 individual runs, using three spiked levels and one nonspiked level of nonsmoker urine samples. The use of spiked urine samples to determine analyte LoDs allowed us to account for sample matrix effects directly, in contrast to other methods (4, 35, 37, 40), where nonmatrix standard(s) and information from the calibration curve were used to estimate the LoD. In obtaining relatively low analyte LoDs (particularly for 1AMN, 2AMN and 4ABP), we are able to apply our method to detect the candidate AAs at trace levels (ppb to ppt). Furthermore, as we have optimized our method toward using lower sample volumes—2 mL, compared to reported ranges of 4–20 mL (4, 5, 35–44)—the method becomes more practicable for use in epidemiological studies and large surveys such as the NHANES. Typically, 30–60-m long columns, coated with phenyl- or methyl-based stationary phases (midpolar to nonpolar range, respectively), have been used for AA analysis (5, 33, 36, 37, 39). Our assay was previously developed and validated using the midpolar Agilent J&W DB-17MS column (30 m × 0.25 mm × 0.25 μm), which is composed of a (50%-phenyl)-methylpolysiloxane stationary phase (44). However, the highly polar DB-FFAP column proved to be much more rugged and specific to the analytes in the assay panel. Specifically, when analyzing urine samples with the DB-17MS column, 26DM’s quantitation ion peak almost exclusively coeluted with an interfering peak, making accurate detection and quantification of this analyte to be particularly difficult. As shown in Supplementary Figure S1, with the use of the DB-FFAP column, analytical resolution was greatly enhanced for the 26DM analyte. Furthermore, when operating with the DB-FFAP column, we were able to reduce the LoD of 4ABP in urine samples from approximately 9.0 to 1.8 ng/L. The marked improvement in 4ABP’s detection rate (with both the quantitation and confirmation ions) was largely due to the lower level of column bleed exhibited by the DB-FFAP column, at the operational temperature utilized. With better detection of both the quantitation and confirmation ions, we were more confident in identifying the 4ABP analyte from a complex matrix such as urine. For the first time, we report the use of SLE, Isolute™ cartridges in sample cleanup of urinary AAs following hydrolysis. These cartridges have a hydrophilic frit that enables water and other aqueous components of the urine samples to absorb onto the bedding material quickly, followed by extraction of the AAs with DCM. The SLE replaced multiple steps used in our previous sample pretreatment process (44) that involved liquid–liquid extraction with hexane, back extraction with 0.1 M HCl and cleanup with a hydrophilic–lipophilic-balanced cartridge. The use of SLE cartridges can also potentially replace other labor-intensive sample cleanup steps reported in the literature (4, 5, 35–43). The automated sample data flow reduces potential human error in sample tracking, handling and data analysis. All relevant sample information, such as the sample ID, sample volume and any dilution factors are saved with each raw data file as the samples are analyzed on the GC–MS/MS. Our current method setup allows for one of two data analyses processes to occur: one, through the Agilent MassHunter Quantitative Analysis software; and the other was through Indigo Biosystems ASCENT™ platform. Although results obtained from either software platforms were proven to be statistically equivalent, we currently use Indigo ASCENT™, exclusively, to analyze all sample data. Through the use of Indigo ASCENT, numerous custom QA rules could be imbedded to ensure consistent evaluation of each analyte on a sample-to-sample, batch-by-batch basis. Conclusion Overall, the new automated method is time-efficient and precise. At least two batches of 32 samples can be prepared each day when higher sample throughput is needed (compared to 1 batch of 43 samples). Laboratory automation has improved the overall accuracy and precision of the analytical method, whereas the newly implemented sample data flow has improved sample tracking and data analysis, including evaluation of individual sample QA and run QC. The complete sample data flow, from sample receiving to final result reporting, is efficient and minimizes potential source(s) of human errors. Acknowledgments This work was funded by the US Food and Drug Administration and the Centers for Disease Control and Prevention, Department of Health and Human Services. The authors would like to acknowledge Sam P. Caudill, Rey Decastro and Yao Li for their help with statistical analyses. The authors would like to acknowledge former team members Elizabeth Cowan, Susan Pyatt, Aaron Danberry, Kyle Emer and James Alex Hodgson for their help. References 1 IARC . ( 1986 ) Tobacco Smoking . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 38 , 1 – 432 . OpenURL Placeholder Text WorldCat 2 Silverman , D. T. , Hartge , P., Morrison , A. S., Devesa , S. S. ( 1992 ) Epidemiology of bladder cancer . Hematology/Oncology Clinics of North America , 6 , 1 – 30 . Google Scholar Crossref Search ADS PubMed WorldCat 3 Hammond , S. K. , Coghlin , J., Gann , P. H., Paul , M., Taghizadeh , K., Skipper , P. L., Tannenbaum , S. R. ( 1993 ) Relationship between environmental tobacco smoke exposure and carcinogen-hemoglobin adduct levels in nonsmokers . Journal of the National Cancer Institute , 85 , 474 – 477 . Google Scholar Crossref Search ADS PubMed WorldCat 4 Teass , A. W. , DeBord , D. G., Brown , K. K., Cheever , L. E., Stettler , L. E., Savage , R. E., Weigel , W. W., Dankovic , D., Ward , E. ( 1993 ) Biological monitoring for occupational exposures to o-toluidine and aniline . International Archives of Occupational and Environmental Health , 65 , S115 – 118 . Google Scholar Crossref Search ADS PubMed WorldCat 5 Riffelmann , M. , Muller , G., Schmieding , W., Popp , W., Norpoth , K. ( 1995 ) Biomonitoring of urinary aromatic amines and arylamine hemoglobin adducts in exposed workers and nonexposed control persons . International Archives of Occupational and Environmental Health , 68 , 36 – 43 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Tang , D. , Warburton , D., Tannenbaum , S. R., Skipper , P. L., Santella , R. M., Cerijido , G. S., Crawford , F. G., Perera , F. P. ( 1999 ) Molecular and Genetic Damage from Environmental Tobacco Smoke in Young Children . Cancer Epidemiology, Biomakers & Prevention , 8 , 427 – 431 . Google Scholar OpenURL Placeholder Text WorldCat 7 Skipper , P. L. , Tannenbaum , S. R., Ross , R. K., Yu , M. C. ( 2003 ) Nonsmoking-related Arylamine Exposure and Bladder Cancer Risk . Cancer Epidemiology, Biomakers & Prevention , 12 , 503 – 507 . Google Scholar OpenURL Placeholder Text WorldCat 8 IARC . ( 2004 ) Tobacco Smoke and Involuntary Smoking . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 83 , 1 – 1473 . PubMed OpenURL Placeholder Text WorldCat 9 USHHS . The Health Consequences of Involuntary Exposure to Tobacco Smoke: A Report of the Surgeon General–Executive Summary . U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, Coordinating Center for Health Promotion, National Center for Chronic Disease Prevention and Health Promotion, Office on Smoking and Heath : Atlanta, GA , 2006 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 10 Bryant , M. S. , Vineis , P., Skipper , P. L., Tannenbaum , S. R. ( 1988 ) Hemoglobin adducts of aromatic amines: Associations with smoking status and type of tobacco . Proceedings of the National Academy of Sciences , 85 , 9788 – 9791 . Google Scholar Crossref Search ADS WorldCat 11 Castelao , J. E. , Yuan , J.-M., Skipper , P. L., Tannenbaum , S. R., Gago-Dominguez , M., Crowder , J. S., Ross , R. K., Yu , M. C. ( 2001 ) Gender- and Smoking-Related Bladder Cancer Risk . Journal of the National Cancer Institute , 93 , 538 – 545 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Vineis , P. ( 1994 ) Epidemiology of cancer from exposure to arylamines . Environmental Health Perspectives , 102 , 7 – 10 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 13 IARC . ( 2010 ) Some aromatic amines, organic dyes, and related exposures . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 99 , 1 – 678 . PubMed OpenURL Placeholder Text WorldCat 14 Stabbert , R. , Schäfer , K. H., Biefel , C., Rustemeier , K. ( 2003 ) Analysis of aromatic amines in cigarette smoke . Rapid Communications in Mass Spectrometry , 17 , 2125 – 2132 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Fischer , P. Chapter 11A. Tobacco Blending. In: Davis D.L., Nielson M.T. (eds). Tobacco: Production, Chemistry and Technology . Oxford; Malden, MA : Blackwell Science , 1999 ; pp. 348. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 16 Dawson , R. F. ( 1952 ) Chemistry and biochemistry of green tobacco . Industrial and Engineering Chemistry , 44 , 266 – 270 . Google Scholar Crossref Search ADS WorldCat 17 Patrianakos , C. , Hoffmann , D. ( 1979 ) Chemical studies on tobacco smoke LXIV. On the analysis of aromatic amines in cigarette smoke . Journal of Anlaytical Toxicology , 3 , 150 – 154 . Google Scholar Crossref Search ADS WorldCat 18 Schmeltz , I. , Hoffmann , D. ( 1977 ) Nitrogen-containing compounds in tobacco and tobacco smoke . Chemical Reviews , 77 , 295 – 311 . Google Scholar Crossref Search ADS WorldCat 19 Heckman , R. A. , Best , F. W. ( 1981 ) An investigation of the lipophilic bases of cigarette smoke condensate . Tobacco Science , 25 , 33 – 39 . Google Scholar OpenURL Placeholder Text WorldCat 20 IARC . ( 2000 ) ortho-TOLUIDINE . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 77 , 267 – 322 . PubMed OpenURL Placeholder Text WorldCat 21 IARC . ( 1993 ) 2,6-DIMETHYLANILINE (2,6-XYLIDINE) . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 57 , 323 – 335 . PubMed OpenURL Placeholder Text WorldCat 22 IARC . ( 2010 ) 4-AMINOBIPHENYL . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 99 , 71 – 109 . OpenURL Placeholder Text WorldCat 23 IARC . ( 2010 ) 2-NAPHTHYLANIME . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 99 , 369 – 405 . OpenURL Placeholder Text WorldCat 24 IARC . ( 1999 ) ortho-ANISIDINE . IARC Monographs on the Evaluation of Carcinogenic Risks to Humans , 73 , 49 – 58 . PubMed OpenURL Placeholder Text WorldCat 25 DeBruin , L. S. , Pawliszyn , J. B., Josephy , P. D. ( 1999 ) Detection of monocyclic aromatic amines, possible mammary carcinogens, in human milk . Chemical Research in Toxicology , 12 , 78 – 82 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Pereira , L. , Mondal , P. K., Alves , M. Aromatic amines sources, environmental impact and remediation. In: Lichtfouse , E., Schwarzbauer , J., Robert , D. (eds.). Pollutants in Buildings, Water and Living Organisms . Cham, Switzerland : Springer , 2015 ; pp. 297 – 346 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 27 Vitzthum , O. G. , Werkhoff , P., Hubert , P. ( 1975 ) New volatile constituents of black tea aroma . Journal of Agricultural and Food Chemistry , 23 , 999 – 1003 . Google Scholar Crossref Search ADS WorldCat 28 Neurath , G. B. , Dünger , M., Pein , F. G., Ambrosius , D., Schreiber , O. ( 1977 ) Primary and secondary amines in the human environment . Food and Cosmetics Toxicology , 15 , 275 – 282 . Google Scholar Crossref Search ADS PubMed WorldCat 29 Chiang , T. A. , Pei-Fen , W., Ying , L. S., Wang , L. F., Ko , Y. C. ( 1999 ) Mutagenicity and aromatic amine content of fumes from heated cooking oils produced in Taiwan . Food and Chemical Toxicology , 37 , 125 – 134 . Google Scholar Crossref Search ADS PubMed WorldCat 30 Bartsch , H. , Caporaso , N., Coda , M., Kadlubar , F., Malaveille , C., Skipper , P. L., Talaska , G., Tannenbaum , S. R., Vineis , P. ( 1990 ) Carcinogen hemoglobin adducts, urinary mutagenicity, and metabolic phenotype in active and passive cigarette smokers . Journal of the National Cancer Institute , 82 , 1826 – 1831 . Google Scholar Crossref Search ADS PubMed WorldCat 31 Bryant , M. S. , Skipper , P. L., Tannenbaum , S. R., Maclure , M. ( 1987 ) Hemoglobin adducts of 4-aminobiphenyl in smokers and nonsmokers . Cancer Research , 47 , 602 – 208 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 32 USHHS . Reporting Harmful and Potentially Harmful Constituents in Tobacco Products and Tobacco Smoke Under Section 904(a)(3) of the Federal Food, Drug, and Cosmetic Act . U.S. Department of Health and Human Services, Food and Drug Administration, Center for Tobacco Products (CTP) : Silver Spring, MD , 2012 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 33 Riedel , K. , Scherer , G., Engl , J., Hagedorn , H. W., Tricker , A. R. ( 2006 ) Determination of three carcinogenic aromatic amines in urines of smokers and nonsmokers . Journal of Analytical Toxicology , 30 , 187 – 195 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Talaska , G. , Zoughool , M. A. ( 2003 ) Aromatic amine and biomarkers of human exposure . Journal of Environmental Science and Health. Part C, Environmental Carcinogenesis & Ecotoxicology Reviews , 21 , 133 – 164 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Amiri , A. , Baghayeri , M., Nori , S. ( 2015 ) Magnetic solid-phase extraction using poly(para-phenylenediamine) modified with magnetic nanoparticles as adsorbent for analysis of monocyclic aromatic amines in water and urine samples . Journal of Chromatography. A , 1415 , 20 – 26 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Labat , L. , Thomas , J., Dehon , B., Humbert , L., Leleu , B., Nisse , C., Lhermitte , M. ( 2006 ) Evaluation of a profession exposition of ortho-toluidine by gas phase chromatography coupled with mass spectrometry . Acta Clinica Belgica , 61 , 63 – 67 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Lamani , X. , Horst , S., Zimmermann , T., Schmidt , T. C. ( 2015 ) Determination of aromatic amines in human urine using comprehensive multi-dimensional gas chromatography mass spectrometry (GCxGC-qMS) . Analytical and Bioanalytical Chemistry , 407 , 241 – 252 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Muz , M. , Ost , N., Kühne , R., Schüürmann , G., Brack , W., Krauss , M. ( 2017 ) Nontargeted detection and identification of (aromatic) amines in environmental samples based on diagnostic derivatization and LC-high resolution mass spectrometry . Chemosphere , 166 , 300 – 310 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Weiss , T. , Angerer , J. ( 2002 ) Simultaneous determination of various aromatic amines and metabolites of aromatic nitro compounds in urine for low level exposure using gas chromatography-mass spectrometry . Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences , 778 , 179 – 192 . Google Scholar Crossref Search ADS WorldCat 40 Yu , J. , Wang , S., Zhao , G., Wang , B., Ding , L., Zhang , X., Xie , J., Xie , F. ( 2014 ) Determination of urinary aromatic amines in smokers and nonsmokers using a MIPs-SPE coupled with LC-MS/MS method . Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences , 958 , 130 – 135 . Google Scholar Crossref Search ADS WorldCat 41 Jiang , C. , Sun , Y., Yu , X., Gao , Y., Zhang , L., Wang , Y., Zhang , H., Song , D. ( 2014 ) Application of C18-functional magnetic nanoparticles for extraction of aromatic amines from human urine . Journal of Chromatography B: Analytical Technologies in the Biomedical and Life Sciences , 947–948 , 49 – 56 . Google Scholar Crossref Search ADS WorldCat 42 DeBruin , L. S. , Josephy , P. D., Pawliszyn , J. B. ( 1998 ) Solid-phase microextraction of monocyclic aromatic amines from biological fluids . Analytical Chemistry , 70 , 1986 – 1992 . Google Scholar Crossref Search ADS PubMed WorldCat 43 Grimmer , G. , Dettbarn , G., Seidel , A., Jacob , J. ( 2000 ) Detection of carcinogenic aromatic amines in the urine of non-smokers . Science of the Total Environment , 247 , 81 – 90 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Seyler , T. H. , Bernert , J. T. ( 2011 ) Analysis of 4-aminobiphenyl in smoker’s and nonsmoker’s urine by tandem mass spectrometry . Biomarkers: Biochemical Indicators of Exposure, Response, and Susceptibility to Chemicals , 16 , 212 – 221 . Google Scholar Crossref Search ADS PubMed WorldCat 45 Caudill , S. P. , Schleicher , R. L., Pirkle , J. L. ( 2008 ) Multi-rule quality control for the age-related eye disease study . Statistics in Medicine , 27 , 4094 – 4106 . Google Scholar Crossref Search ADS PubMed WorldCat 46 CLSI . ( 2004 ) Protocols for determination of limits of detection and limits of quantitation: approved guideline . CLSI document EP17-A , 24 , 1 – 52 . OpenURL Placeholder Text WorldCat Author notes The findings and conclusions in this report are those of the author(s) and do not necessarily represent the official position of the Centers for Disease Control and Prevention. Published by Oxford University Press 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US. Published by Oxford University Press 2018.

Collins, B J; Slade, D; Ryan, K; Mathias, R; Shan, A; Algaier, J; Aillon, K; Waidyanatha, S

doi: 10.1093/jat/bky048pmid: 30060005

Abstract Tris(chloropropyl)phosphate (TCPP) is an organophosphorus flame retardant (OPFR) and plasticizer increasingly used in consumer products and as a replacement for brominated flame retardants. Commercially available TCPP is a mixture of four structural isomers the most abundant of which is tris(1-chloro-2-propyl)phosphate (TCPP-1). Although there is a widespread use of TCPP and potential for human exposure, there is limited data on the safety or toxicity of TCPP. The National Toxicology Program is conducting long-term studies to examine the toxicity of the TCPP in rats after lifetime exposure, including perinatal oral exposure. Quantitative estimates of internal dose are essential to interpret toxicological findings in rodents. To aid in this, a method was fully validated to quantitate the most abundant isomer, TCPP-1, in female Harlan Sprague Dawley (HSD) rat and B6C3F1 mouse plasma with partial validation in male rat plasma, and male and female mouse plasma. The method used protein precipitation using trichloroacetic acid followed by the extraction with toluene, and analysis by gas chromatography with flame photometric detection. The performance of the method was evaluated over 5–70 ng TCPP-1/mL plasma. The method was linear (r ≥ 0.99), accurate (inter-day relative error: ≤ ± –7.2) and precise (inter-batch relative standard deviation: ≤27.5%). The validated method has lower limits of quantitation and detection of ~5 and 0.9 ng/mL, respectively, in female HSD rat plasma and can be used on samples as small as 50 μL demonstrating the applicability to plasma samples from toxicology studies. Introduction Organophosphorus chemicals (OPCs) are widely used in commercial applications as flame retardants (FR), plasticizers, insulators and viscosity regulators. Tris(chloropropyl)phosphate (TCPP) is an organophosphorus flame retardant (OPFR) and plasticizer increasingly used in consumer products, including polyurethane, and as a replacement for brominated FRs and other toxic chlorinated FRs (1, 2). Commercial TCPP is a mixture of four isomers: tris(1-chloro-2-propyl) phosphate (TCPP-1), bis(2-chloro-1-methylethyl) 2-chloropropyl phosphate (TCPP-2), bis(2-chloropropyl) 2-chloroispropyl phosphate (TCPP-3), and tris(2-chloropropyl) phosphate (TCPP-4) (Figure 1) and the mixture is often referenced by its most abundant isomer, TCPP-1; TCPP-1 comprises 50–85% of commercial TCPP, followed by TCPP-2 (15–40%), TCPP-3 (<15%), and TCPP-4 (<1%) (3). In 2012, the National Production Volume of TCPP was ~54 million pounds (4). Due to its commercial use and physiochemical properties, TCPP is now considered “ubiquitous” in the environment and has been detected in airborne particulates (5, 6), estuarine systems (7), as well as in household dust and indoor air (8–15), home furniture (16), baby products are (17), daycare centers (18, 19) and elementary schools (10, 12, 19). Furthermore, TCPP has been detected (0.2–17 μg/kg) in marine food sources such as fish, crab and mussels which highlights an additional source for human exposure (20–23). Van der Veen and de Boer have summarized these observations and concentrations (ng–mg) of TCPP found in various environmental sources (2). TCPP has also been measured in human subjects’ blood, breast milk and urine (23–26), which links environmental exposures with human exposure via oral, dermal or inhalation routes. Figure 1. Open in new tabDownload slide Tris(chloropropyl) phosphate (TCPP) isomers. (A) tris(1-chloro-2-propyl) phosphate (TCPP-1). (B) bis(2-chloro-1-methylethyl) 2-chloropropyl phosphate (TCPP-2). (C) bis(2-chloropropyl) 2-chloroisopropyl phosphate (TCPP-3). (D) tris(2-chloropropyl) phosphate (TCPP-4). TCPP was nominated to the National Toxicology Program (NTP) for toxicity testing based on limited toxicity data available and anticipated increased use in consumer products. The bulk of the available TCPP toxicity information is summarized from industry studies and is reported in risk assessment documents and chemical dossiers (3, 27–30). Unfortunately, many of these studies are not publicly available for full review. Existing evidence demonstrates that TCPP is non-genotoxic (28, 29, 31) and has minimal in vivo toxicity with the exception that kidney and liver appear to be target organs following short-term repeat oral exposure (32). There have been no studies completed evaluating the carcinogenic potential of chronic TCPP exposure. In vitro, TCPP is active in assays for endocrine disruption in H295R cells (31), results in gene expression changes associated with metabolism, lipid regulation and growth in hepatocytes and neuronal cells (33) and decreases cell number and differentiation in neuronal PC12 cells (32). As the toxicity database for TCPP continues to grow, there is a need for more definitive assessments of its potential effects on reproduction and development as well as carcinogenicity, which the NTP toxicity studies of TCPP were conducted to address. Quantitative estimates of internal dose after TCPP exposure are essential to interpret toxicological findings in rodents. Analytical methods to measure TCPP have been developed for surface waters (34), milk powder (35), human hair (36), blood and urine (24–26, 37, 38). Zhao et al. (24) developed a method using ultraperformance liquid chromatography-tandem mass spectrometry (UPLC–MS-MS) to measure a suite of OPCs, including TCPP, in human whole blood with a limit of detection (LOD) of 0.19 ng/mL. The method employed extensive sample cleanup and concentration steps to achieve the low detection limits. Jonsson and Nilsson (39) developed a method using gas chromatography-nitrogen phosphorus detection (GC-NPD) to analyze human plasma for TCPP with a TCPP detection limit of ~0.5 ng/mL; however, this method also employs extensive, specialized sample cleanup and concentration steps. The objective of our work was to develop and validate a simple, efficient and sensitive method to quantitate TCPP in NTP studies without the need for extensive and/or specialized (e.g., microfiltration) cleanup, which can be labor-intensive. Our method specifically quantitates TCPP-1 in plasma, as a marker of internal exposure in TCPP-exposed Harlan Sprague Dawley (HSD) rats and B6C3F1 mice using GC with flame photometric detection (FPD). The method demonstrates the efficiency and sensitivity required, with minimal sample cleanup, to apply to biological matrices from the toxicology study. Materials and Methods Chemicals and reagents TCPP was purchased in two lots from Albemarle Corporation (Baton Rouge, LA) and was combined to produce one lot. The identity of the material as TCPP was confirmed by infrared and nuclear magnetic resonance spectroscopy. Purity of the TCPP was determined using gas chromatography with flame ionization detection (GC-FID) (Agilent, Santa Clara, CA); four major peaks and eight minor peaks were present in the sample. The four major peaks were identified as TCPP isomers by GC–MS with TCPP-1 constituting 67.54% of the material, followed by TCPP-2 (25.67%), TCPP-3 (3.63%) and TCPP-4 (0.22%). The eight minor peaks were impurities totaling 2.73%. The purity of the material, determined by summing the confirmed TCPP isomers, was 97.06%. The internal standard (IS), tripentyl phosphate (TPP), was purchased from Tokyo Chemical Industry Co., Ltd (Portland, OR). Ethanol (Koptec, 95%) was purchased from VWR (Radnor, PA); 200 proof ethanol from Pharmco-AAPER (Shelbyville, KY); toluene (99.9%) from Burdick & Jackson (Muskegon, MI); hexane (99.99%) from Honeywell (Morris Plains, NJ); trichloroacetic acid (TCA, 0.6N solution, 10% w/v); acetone (≥99.9%); ethyl acetate (≥99.7%) from Sigma-Aldrich (St. Louis, MO) and acetonitrile (99.9%) from Fisher Scientific (Waltham, MA). Control HSD rat (male PND 28 pup, adult female) and B6C3F1 mouse (adult male and female) plasma with sodium heparin (Na-heparin), disodium-EDTA (Na2-EDTA) or tripotassium-EDTA (K3-EDTA) anticoagulant was purchased from BioreclamationIVT, LLC (New York, NY). Experimental design TCPP was used to prepare all samples and standards used in the validation and sample analysis. Because the TCPP-1 isomer was the only analyte quantitated, all reported concentrations are for TCPP-1, which were calculated from the determined concentration of TCPP-1 in TCPP (67.54%). Linearity and calibration range, precision (relative standard deviation, RSD) and intra- and inter-day accuracy (relative error, RE), reproducibility, measurement limits or sensitivity (lower limit of quantitation [LLOQ], defined as the lowest standard with acceptable precision and accuracy [RE ≤ 20%; RSD ≤ 20%]; limit of detection [LOD], calculated as three times the standard deviation of the determined concentrations for the replicate preparations of the lowest spiked matrix standard), selectivity, dilution verification and recovery were evaluated for the analytical method in female HSD plasma using spiked matrix standard calibration curves prepared at eight concentrations over the range of 5–70 ng/mL and analyzed on four separate analysis days. Selectivity was evaluated in six independent lots of adult female HSD plasma. A matrix factor was calculated by dividing the mean response for the IS matrix blank replicates (n = 3) by the mean response for the LLOQ (5 ng/mL) matrix standard replicates (n = 6) for each plasma lot. Dilution verification was conducted to evaluate whether samples above the upper limit of quantitation (ULOQ) of the validated range could be accurately quantified after diluting into the range. Reproducibility was evaluated using a split matrix calibration curve prepared in 200-μL aliquots of adult female HSD rat plasma (n = 14). The final reconstituted extracts for each matrix calibration standard were divided into two aliquots, which were analyzed on the same day on the same instrument, at the beginning and end of sample batch. Results were reported as relative response factor (%RRF), a comparison of the matrix standard peak area ratio (PAR) normalized as a percentage of the expected concentration for the standards analyzed at each time. Instrument ruggedness was evaluated using a split matrix calibration curve prepared in 200-μL aliquots of female rat plasma (n = 14). The final reconstituted extracts for each matrix calibration standard were divided into two aliquots, which were analyzed on the same day on two separate instruments using identical analytical conditions. Results were reported as %RRF. The method was partially validated in male PND 28 rat pup plasma as well as adult male and female B6C3F1 mouse plasma over the range of 5–70 ng/mL. Linearity, sensitivity, precision, accuracy, extraction efficiency (recovery) and selectivity were evaluated using blanks and standard curves prepared in each secondary matrix. The effect of sample volume was evaluated by preparing spiked plasma standard curves at eight concentrations over the range of 5–70 ng/mL in 100- or 200-μL aliquots of female HSD rat plasma, and by spiking 50- or 100-μL aliquots of female B6C3F1 mouse plasma with TCPP-1 at 10 or 60 ng/mL. The effect of the anticoagulant used was evaluated by spiking 100-μL aliquots of female HSD rat plasma containing Na-EDTA, Na2-EDTA or K3-EDTA with TCPP-1 at 10 or 50 ng/mL. Spiked samples were processed and analyzed as described for female HSD rat plasma. Stability of female HSD plasma extracts over 7 days at ambient or refrigerator (1–5˚C) temperatures and stability of spiked plasma standards after three freeze–thaw cycles over 14 days were evaluated using replicate matrix standards prepared at 10, 30 and 50 ng/mL. To evaluate storage stability, replicate, spiked female HSD plasma samples were prepared at 10 and 50 ng/mL and stored under freezer (−20°C) or ultracold (−80°C) for up to 60 days. Sample preparation and analysis Two TCPP stock solutions were prepared at ~1 and ~0.7 mg/mL by spiking specific volumes of TCPP into ethanol. Intermediate stocks and spiking solutions were prepared by diluting the TCPP stocks with ethanol. Reported concentrations are for TCPP-1 and were based on 67.54% TCPP-1 found to be present in the TCPP lot used. Henceforth we will refer to TCPP-1 when describing concentration in standards and samples. TCPP-1 solvent standard curves were prepared in toluene at eight concentrations over the range of 5–70 ng/mL (n = 1) using aliquots of alternating intermediate stocks. TCPP-1 matrix standard curves at eight concentrations (n = 3 at low, medium and high; rest n = 1) were prepared over the concentration range of 5–70 ng/mL by spiking 200-μL plasma aliquots with 20 μL of an appropriate TCPP-1 spiking solution, followed by the addition of TCA (40 μL) and vortex mixing for ~30 s. The standards were left undisturbed for ~15 min and extracted twice with 1 mL of toluene by vigorous vortex mixing for ~1 min, and centrifugation for ~5 min at ~2,800× g. The supernatants from the two extractions were combined and evaporated to dryness under nitrogen at 25°C. The residue was reconstituted with 200 μL of IS working solution. TCPP-1 matrix quality control (QC) standards were prepared at 5 (low), 20 (medium) and 70 (high) ng/mL. Stability QC samples were prepared at 10 or 50 ng/mL using ~20 mL of pooled control adult female HSD plasma. Stock solutions of TCPP were newly prepared for each analysis day. All solutions and standards were prepared in glassware prewashed with acetone in order to minimize contamination from environmentally occurring TCPP (40, 41). IS solution was prepared by dissolving TPP in toluene (working solution: 40 ng/mL). Stock solutions of TCPP were freshly prepared on each analysis day. The assessment of the stability of the TPP (IS) stock solution at ambient temperature for over a period of 2 weeks showed no degradation. Samples were analyzed by GC-FPD. The analytical system consisted of an Agilent (Santa Clara, CA) 6890 GC-FPD with a Restek (Bellefonte, PA) Rtx-5MS column (30 m × 0.25 mm × 1.0 μm) and a Restek Integra-Guard guard column (10 m). Sample extract (3 μL) was introduced onto the column via splitless injection at 250°C. Analyte separation was achieved using helium carrier gas at ~2.0 mL/min with a gradient temperature program starting at 125°C for 2 min, 20°C/min to 215°C held for 3 min, 1.5°C/min to 222°C held for 1.3 min and 30°C/min to 340°C held for 19 min. Detection was performed at 250°C with a phosphorus filter. In phosphorus mode, the characteristic chemiluminescence of excited hydrogen phosphorus oxide (HPO*) is measured at 526 nm, which results in a linear response for emission intensity over a wide concentration range (42). Hydrogen, oxidizer (air) and make-up gas (He) flow rates were 75, 100 and 16 mL/min, respectively. The combined detector make-up flow (He) was 18 mL/min. The retention times of TCPP-1 and IS were ~13.5 and ~16.0 min, respectively. Quantitation was performed using PAR (TCPP-1/IS) plotted against plasma TCPP-1 concentration (ng/mL). The slope, intercept and correlation coefficient (r) of the best-fit linear regression were calculated from a weighted (1/x) least squares linear regression analysis. Results Validation in female rat plasma A GC-FPD method for the determination of TCPP-1 in HSD female rat plasma was successfully validated and partial validations showed that the method could also be used to quantitate TCPP-1 in PND 28 HSD male rat pup plasma as well as in male and female B6C3F1 mouse plasma. System suitability was evaluated at the beginning of each analysis. The analytical method was shown to be linear over the range of 5–70 ng/mL (r ≥ 0.99); representative matrix and solvent calibration curves are shown in Figure 2. Typical chromatograms of a matrix standard and a matrix blank are shown in Figure 3. The LLOQ and LOD were ~5 and ~1 ng/mL, respectively (Table I). Table I. Analytical Method Validation Data for Male and Female Rat and Mouse Plasma Parameter . Female rat . Male rat pup . Female mouse . Male mouse . Validation scope . Full . Partial . Partial . Partial . LOD (ng/mL) 1 1 0.3 1 LLOQ (ng/mL) 5 5 5 5 Range (ng/mL) 5–70 5–70 5–70 5–70 r-Value ≥0.99 ≥0.99 ≥0.99 ≥0.99 Mean recovery (%RSD) 85.5 (6.0) 100.4 (8.8) 87.6 (7.1) 86.1 (15.4) Intra-day precision (%RSD) Range Mean Mean Mean Low QC (5 ng/mL) 6.3–46.8 26.2 44.1 42.6 Mid QC (20 ng/mL) 2.7–9.1 13.6 9.3 5.6 High QC (70 ng/mL) 1.7–9.3 2.9 11.2 9.4 Intra-day accuracy (%RE) Range Mean Mean Mean Low QC (5 ng/mL) −33.4–13.7 24.3 51.4 33.9 Mid QC (20 ng/mL) 1.3–12.2 −7.1 0.1 7.7 High QC (70 ng/mL) −3.0–0.7 −0.9 −6.7 −2.0 Inter-day precision (%RSD) Mean – – – Low QC (5 ng/mL) 27.5 ND ND ND Mid QC (20 ng/mL) 7.3 ND ND ND High QC (70 ng/mL) 5.4 ND ND ND Inter-day accuracy (%RE) Mean – – – Low QC (5 ng/mL) −7.2 ND ND ND Mid QC (20 ng/mL) 5.1 ND ND ND High QC (70 ng/mL) −1.3 ND ND ND Selectivity (Matrix Factor, %) 10.7 1.6 16.9 43.2 Instrument ruggednessa (%RRF) Mean (Std. Dev.) – – – Instrument 1 2.58 (0.72) ND ND ND Instrument 2 2.35 (0.70) ND ND ND Parameter . Female rat . Male rat pup . Female mouse . Male mouse . Validation scope . Full . Partial . Partial . Partial . LOD (ng/mL) 1 1 0.3 1 LLOQ (ng/mL) 5 5 5 5 Range (ng/mL) 5–70 5–70 5–70 5–70 r-Value ≥0.99 ≥0.99 ≥0.99 ≥0.99 Mean recovery (%RSD) 85.5 (6.0) 100.4 (8.8) 87.6 (7.1) 86.1 (15.4) Intra-day precision (%RSD) Range Mean Mean Mean Low QC (5 ng/mL) 6.3–46.8 26.2 44.1 42.6 Mid QC (20 ng/mL) 2.7–9.1 13.6 9.3 5.6 High QC (70 ng/mL) 1.7–9.3 2.9 11.2 9.4 Intra-day accuracy (%RE) Range Mean Mean Mean Low QC (5 ng/mL) −33.4–13.7 24.3 51.4 33.9 Mid QC (20 ng/mL) 1.3–12.2 −7.1 0.1 7.7 High QC (70 ng/mL) −3.0–0.7 −0.9 −6.7 −2.0 Inter-day precision (%RSD) Mean – – – Low QC (5 ng/mL) 27.5 ND ND ND Mid QC (20 ng/mL) 7.3 ND ND ND High QC (70 ng/mL) 5.4 ND ND ND Inter-day accuracy (%RE) Mean – – – Low QC (5 ng/mL) −7.2 ND ND ND Mid QC (20 ng/mL) 5.1 ND ND ND High QC (70 ng/mL) −1.3 ND ND ND Selectivity (Matrix Factor, %) 10.7 1.6 16.9 43.2 Instrument ruggednessa (%RRF) Mean (Std. Dev.) – – – Instrument 1 2.58 (0.72) ND ND ND Instrument 2 2.35 (0.70) ND ND ND Std. Dev., standard deviation; ND, not determined. aSystem 1 and System 2: 6890 GC-FPD (Agilent, Santa Clara, CA); Rtx-5MS column (30 m × 0.25 mm × 1.0 μm); Integra-Guard guard column (10 m) (Restek Bellefonte, PA) Open in new tab Table I. Analytical Method Validation Data for Male and Female Rat and Mouse Plasma Parameter . Female rat . Male rat pup . Female mouse . Male mouse . Validation scope . Full . Partial . Partial . Partial . LOD (ng/mL) 1 1 0.3 1 LLOQ (ng/mL) 5 5 5 5 Range (ng/mL) 5–70 5–70 5–70 5–70 r-Value ≥0.99 ≥0.99 ≥0.99 ≥0.99 Mean recovery (%RSD) 85.5 (6.0) 100.4 (8.8) 87.6 (7.1) 86.1 (15.4) Intra-day precision (%RSD) Range Mean Mean Mean Low QC (5 ng/mL) 6.3–46.8 26.2 44.1 42.6 Mid QC (20 ng/mL) 2.7–9.1 13.6 9.3 5.6 High QC (70 ng/mL) 1.7–9.3 2.9 11.2 9.4 Intra-day accuracy (%RE) Range Mean Mean Mean Low QC (5 ng/mL) −33.4–13.7 24.3 51.4 33.9 Mid QC (20 ng/mL) 1.3–12.2 −7.1 0.1 7.7 High QC (70 ng/mL) −3.0–0.7 −0.9 −6.7 −2.0 Inter-day precision (%RSD) Mean – – – Low QC (5 ng/mL) 27.5 ND ND ND Mid QC (20 ng/mL) 7.3 ND ND ND High QC (70 ng/mL) 5.4 ND ND ND Inter-day accuracy (%RE) Mean – – – Low QC (5 ng/mL) −7.2 ND ND ND Mid QC (20 ng/mL) 5.1 ND ND ND High QC (70 ng/mL) −1.3 ND ND ND Selectivity (Matrix Factor, %) 10.7 1.6 16.9 43.2 Instrument ruggednessa (%RRF) Mean (Std. Dev.) – – – Instrument 1 2.58 (0.72) ND ND ND Instrument 2 2.35 (0.70) ND ND ND Parameter . Female rat . Male rat pup . Female mouse . Male mouse . Validation scope . Full . Partial . Partial . Partial . LOD (ng/mL) 1 1 0.3 1 LLOQ (ng/mL) 5 5 5 5 Range (ng/mL) 5–70 5–70 5–70 5–70 r-Value ≥0.99 ≥0.99 ≥0.99 ≥0.99 Mean recovery (%RSD) 85.5 (6.0) 100.4 (8.8) 87.6 (7.1) 86.1 (15.4) Intra-day precision (%RSD) Range Mean Mean Mean Low QC (5 ng/mL) 6.3–46.8 26.2 44.1 42.6 Mid QC (20 ng/mL) 2.7–9.1 13.6 9.3 5.6 High QC (70 ng/mL) 1.7–9.3 2.9 11.2 9.4 Intra-day accuracy (%RE) Range Mean Mean Mean Low QC (5 ng/mL) −33.4–13.7 24.3 51.4 33.9 Mid QC (20 ng/mL) 1.3–12.2 −7.1 0.1 7.7 High QC (70 ng/mL) −3.0–0.7 −0.9 −6.7 −2.0 Inter-day precision (%RSD) Mean – – – Low QC (5 ng/mL) 27.5 ND ND ND Mid QC (20 ng/mL) 7.3 ND ND ND High QC (70 ng/mL) 5.4 ND ND ND Inter-day accuracy (%RE) Mean – – – Low QC (5 ng/mL) −7.2 ND ND ND Mid QC (20 ng/mL) 5.1 ND ND ND High QC (70 ng/mL) −1.3 ND ND ND Selectivity (Matrix Factor, %) 10.7 1.6 16.9 43.2 Instrument ruggednessa (%RRF) Mean (Std. Dev.) – – – Instrument 1 2.58 (0.72) ND ND ND Instrument 2 2.35 (0.70) ND ND ND Std. Dev., standard deviation; ND, not determined. aSystem 1 and System 2: 6890 GC-FPD (Agilent, Santa Clara, CA); Rtx-5MS column (30 m × 0.25 mm × 1.0 μm); Integra-Guard guard column (10 m) (Restek Bellefonte, PA) Open in new tab Figure 2. Open in new tabDownload slide Typical TCPP-1 matrix and solvent calibration curve for female rat plasma. ▪ Solvent standards and ● matrix standards. Figure 3. Open in new tabDownload slide Typical GC-FPD chromatograms. (A) Female HSD rat plasma matrix standard (50 ng/mL TCPP-1). (B) Matrix blank with IS. For matrix QC samples prepared at 20 and 70 ng/mL TCPP-1, intra- and inter-day precision (RSD) ranged from 1.7% to 9.3% and 5.4% to 7.3%, respectively, and intra- and inter-day accuracy (RE) ranged from −3.0% to 12.2% and −1.3% to 5.1%, respectively (Table I). Inter- and intra-day precision and accuracy at LLOQ (5 ng/mL) showed higher variability, with values ranging from 6.3% to 46.8% (precision) and −33.4% to 13.7% (accuracy) (Table I). Recovery of TCPP-1 in plasma was ≥77.0% over the range of 5–70 ng/mL, with a mean recovery >85%. Selectivity was acceptable with an average matrix factor for the six lots of 10.7% indicating the absence of matrix interferences in female HSD plasma samples (Table I and Supplementary Table S5). Samples (n = 5) were successfully diluted from 175 ng/mL into the calibration range (35 ng/mL) with mean RE and RSD values of −15.4 and 10.4%, respectively. REs for each of the five standards ranged from −24.7% to −5.8%. The reproducibility of the split spiked matrix standard curve was acceptable, with all standards having a relative average deviation (RAD) ≤20%. Instrument ruggedness was acceptable with average %RRF values of 2.58 and 2.35 for instruments 1 and 2, respectively (Table I) and 10 of 14 RAD values ≤ 5.9%. Stability for plasma extracts stored for 7 days was evaluated. The results indicated extracted samples stored under both ambient (light) and refrigerated (dark) conditions are stable (87.6 to 102.5% of Day 0), with no significant difference between the two storage conditions based on comparison of determined concentrations after storage with those calculated on Day 0 (ANOVA [α = 0.05]: P = 0.2910). Partial validation in male HSD rat pup plasma and male and female B6C3F1 mouse plasma Partial validations were conducted to assess linearity, sensitivity, precision and accuracy (intra-day), recovery and selectivity of TCPP-1 in male HSD rat pup (PND 28) plasma, and male and female B6C3F1 mouse plasma over the concentration range of 5–70 ng/mL. PND 28 male HSD pup plasma standard curves were linear (r ≥ 0.99) with LLOQ and LOD values of ~5 and ~1 ng/mL, respectively (Table I). Mean recovery was 100.4% (RSD: 8.8%), and the minimum recovery was 89.9. Selectivity of the method was demonstrated with average matrix factors ranging from 0.0% to 9.8% (n = 3) for six lots of control plasma (Table I and Supplementary Table S5). Intra-day precision (%RSD) at all concentrations was ≤26.2%. Intra-day accuracy (%RE) was −0.9 and −7.1% for the 70 and 20 ng/mL concentrations, respectively. At 5 ng/mL, the mean intra-day accuracy was 24.3% (Table I), however, 3 of the 12 intra-day accuracy standards had REs of >60%, while the remaining standards ranged from −18.4% to 33.6%, with a mean %RE of 8.0%. Female and male B6C3F1 mouse plasma standard curves were linear (r ≥ 0.99) over the range of 5–70 ng/mL with LLOQ and LOD values of ~5 and ≤1 ng/mL, respectively (Table I). Recovery ranged between 79.9% and 104.7%, with mean recoveries of 87.6 and 86.1% for female and male B6C3F1 mice, respectively. Selectivity of the method was demonstrated with average matrix factors ranging from 11.3% to 20.1% (n = 3) for six lots of control female mouse plasma. Average matrix factors were higher for the male mouse plasma ranging from 21.7% to 59.4% (n = 3) for six lots of control plasma (Table I and Supplementary Table S5). For both males and females, intra-day precision (%RSD) at 20 and 70 ng/mL was ≤11.2% and intra-day accuracy (%RE) was ≤±7.7%. As observed for the rats, the 5 ng/mL concentration had higher variability with intra-day precision ≤44.1% and intra-day accuracy ranging from −18.8% to 169% (n = 6); for females, three of the six 5 ng/mL intra-day accuracy standards had REs ≥51%, while the remaining standards ranged from −0.5% to 3.6%, with a mean %RE of 1.5%. For males, two of the six 5 ng/mL intra-day accuracy standards had REs ≥90%, while the remaining standards ranged from −6.2% to 0.6%, with a mean of −2.5%. Plasma volume was determined to have no effect on the quantitation of TCPP-1. In rat plasma, the ANOVA (α = 0.05) comparison of the normalized PAR values at each concentration (%RRF) for the 200 μL vs. 100 μL plasma samples showed that they were statistically equivalent (P ≥ 0.1355) (Supplementary Table S2). In female mouse plasma, 100 μL vs. 50 μL sample volumes were compared at two nominal concentrations (10 or 60 ng/mL). ANOVA (α = 0.05) comparison of the PAR values for the 100 μL vs. 50 μL sample volume at each concentration indicated that they were statistically equivalent (10 ng/mL: P = 0.8911; 60 ng/mL: P = 0.1355) (Supplementary Table S3). The effect of plasma anticoagulant on the analytical method was evaluated by comparing matrix standards prepared using female rat plasma containing Na-heparin, Na2-EDTA or K3-EDTA. Comparison of the PAR values for Na-heparin vs. Na2-EDTA and Na2-EDTA vs. K3-EDTA at two concentrations showed that all types of anticoagulant produced statistically equivalent results (10 ng/mL: P ≥ 0.05022; 60 ng/mL: P ≥ 0.2098). Analyte stability in plasma Replicate TCPP-1 samples prepared at 50 ng/mL in female HSD plasma and stored through three freeze–thaw cycles had an average recovery of 101.8 ± 14.6%. Mean recovery for 30 or 10 ng/mL samples was 75.6 ± 0.8% and 65.0 ± 8.9%, respectively. (Table II). Table II. Sample Extract and Plasma Freeze–Thaw Stability Data. Target concentration (ng/mL) . Mean determined concentrationa (ng/mL) . Precision (%RSD)b . Accuracy (mean %RE)b,c . Stability . % of Day 0 mean (s)b . Precision (%RSD)b . Extracts stored ambient (in light) 50.0 54.12 18.2 7.6 91.7 (16.7) 18.2 30.0 31.28 3.2 3.6 91.7 (3.0) 3.3 10.0 10.90 4.0 10.2 102.5 (4.0) 3.9 Extracts stored refrigerated (in dark) 50.0 55.23 1.0 9.8 93.6 (0.9) 1.0 30.0 29.90 2.2 −1.0 87.6 (2.0) 2.3 10.0 10.01 3.6 1.2 94.2 (3.4) 3.6 Plasma freeze–thaw 50.0 60.06 14.4 19.4 101.8 (14.6) 14.3 30.0 25.79 1.1 −14.6 75.6 (0.8) 1.1 10.0 6.91 13.8 −30.1 65.0 (8.9) 13.7 Target concentration (ng/mL) . Mean determined concentrationa (ng/mL) . Precision (%RSD)b . Accuracy (mean %RE)b,c . Stability . % of Day 0 mean (s)b . Precision (%RSD)b . Extracts stored ambient (in light) 50.0 54.12 18.2 7.6 91.7 (16.7) 18.2 30.0 31.28 3.2 3.6 91.7 (3.0) 3.3 10.0 10.90 4.0 10.2 102.5 (4.0) 3.9 Extracts stored refrigerated (in dark) 50.0 55.23 1.0 9.8 93.6 (0.9) 1.0 30.0 29.90 2.2 −1.0 87.6 (2.0) 2.3 10.0 10.01 3.6 1.2 94.2 (3.4) 3.6 Plasma freeze–thaw 50.0 60.06 14.4 19.4 101.8 (14.6) 14.3 30.0 25.79 1.1 −14.6 75.6 (0.8) 1.1 10.0 6.91 13.8 −30.1 65.0 (8.9) 13.7 aDay 0 determined concentrations: 59.00, 34.11 and 10.63 ng/mL for 50, 30 and 10 ng/mL target concentrations. bn = 4. cCompared to Day 0. Plasma: female HSD rat plasma. Anticoagulant: Na-Heparin. Stability criteria: 75% ≤ determined compared to Day 0 concentration as percentage ≤125%. Open in new tab Table II. Sample Extract and Plasma Freeze–Thaw Stability Data. Target concentration (ng/mL) . Mean determined concentrationa (ng/mL) . Precision (%RSD)b . Accuracy (mean %RE)b,c . Stability . % of Day 0 mean (s)b . Precision (%RSD)b . Extracts stored ambient (in light) 50.0 54.12 18.2 7.6 91.7 (16.7) 18.2 30.0 31.28 3.2 3.6 91.7 (3.0) 3.3 10.0 10.90 4.0 10.2 102.5 (4.0) 3.9 Extracts stored refrigerated (in dark) 50.0 55.23 1.0 9.8 93.6 (0.9) 1.0 30.0 29.90 2.2 −1.0 87.6 (2.0) 2.3 10.0 10.01 3.6 1.2 94.2 (3.4) 3.6 Plasma freeze–thaw 50.0 60.06 14.4 19.4 101.8 (14.6) 14.3 30.0 25.79 1.1 −14.6 75.6 (0.8) 1.1 10.0 6.91 13.8 −30.1 65.0 (8.9) 13.7 Target concentration (ng/mL) . Mean determined concentrationa (ng/mL) . Precision (%RSD)b . Accuracy (mean %RE)b,c . Stability . % of Day 0 mean (s)b . Precision (%RSD)b . Extracts stored ambient (in light) 50.0 54.12 18.2 7.6 91.7 (16.7) 18.2 30.0 31.28 3.2 3.6 91.7 (3.0) 3.3 10.0 10.90 4.0 10.2 102.5 (4.0) 3.9 Extracts stored refrigerated (in dark) 50.0 55.23 1.0 9.8 93.6 (0.9) 1.0 30.0 29.90 2.2 −1.0 87.6 (2.0) 2.3 10.0 10.01 3.6 1.2 94.2 (3.4) 3.6 Plasma freeze–thaw 50.0 60.06 14.4 19.4 101.8 (14.6) 14.3 30.0 25.79 1.1 −14.6 75.6 (0.8) 1.1 10.0 6.91 13.8 −30.1 65.0 (8.9) 13.7 aDay 0 determined concentrations: 59.00, 34.11 and 10.63 ng/mL for 50, 30 and 10 ng/mL target concentrations. bn = 4. cCompared to Day 0. Plasma: female HSD rat plasma. Anticoagulant: Na-Heparin. Stability criteria: 75% ≤ determined compared to Day 0 concentration as percentage ≤125%. Open in new tab Stability of TCPP-1 in adult female HSD plasma was evaluated at 10 and 50 ng/mL following sub-ambient (−20°C or −80°C) storage for up to 60 days. The mean determined concentration on each analysis day was compared to the mean determined concentration on Day 0. Recovery of untreated standards prepared at 50 ng/mL under both storage conditions was ≥89.3 ± 12.6%, with both storage conditions being statistically equivalent (ANVOA [α = 0.05]: P = 0.6759); at 10 ng/mL recovery was ≥49.3 ± 16.2%. Results are shown in Supplementary Table S1. Because recovery of TCPP in plasma was <90% after 60 days, we investigated the effect of treatment of spiked adult female HSD plasma matrix standards (10 or 50 ng/mL) with a precipitation agent (TCA) prior to storage at −80°C for up to 65 days. Pooled aliquots (5 mL) of control plasma were spiked with 50 μL of an appropriate TCPP-1 spiking solution and mixed. A 20-μL aliquot of TCA was added to an aliquot (100 μL) of the pooled spiked plasma, mixed, allowed to stand for 15 min and then stored. Samples were analyzed on the day of preparation and each subsequent analysis day as described for female HSD rat plasma. The mean determined concentration on each analysis day was compared to the mean determined concentration on Day 0. Recovery for treated 50 ng/mL standards stored for 65 days was 113.6 ± 1.0%. Recovery for 10 ng/mL samples was 84.3 ± 0.4% after 65 days and was lower than recovery for 50 ng/mL samples at all intermediate time points. Results are shown in Supplementary Table S1 and Figure 4. Figure 4. Open in new tabDownload slide TCPP-1 stability in plasma. Female HSD rat plasma with and without TCA pretreatment. ♦ 50 ng/mL at −70°C with TCA pretreatment. ▪ 50 ng/mL at −70°C without TCA pretreatment. ● 50 ng/mL at −20°C without TCA pretreatment. ◊ 10 ng/mL at −70°C with TCA pretreatment. □ 10 ng/mL at −70°C without TCA pretreatment. ○ 10 ng/mL at −20°C without TCA pretreatment. Discussion TCPP was nominated to the NTP for toxicity testing due to its increasing use as an FR and a plasticizer in a variety of commercial products and the resulting potential for a widespread human exposure. To assess internal exposure of TCPP following the exposure of rodents via feed in NTP studies, a method for the measurement of TCPP in rodent plasma was developed. TCPP-1 was selected as our analyte because it is the most abundant isomer in TCPP and additional work with plasma from HSD rats and B6C3F1 mice dosed with TCPP (Supplementary Figures S1 and S2 and Table S4) has shown no evidence of significant isomer-specific degradation or interconversion between the isomers, making TCPP-1 a good marker of internal TCPP exposure. Although a couple of analytical methods are available in the literature to quantitate TCPP in blood and urine with detection limits comparable to our method, they require extensive sample preparation and cleanup prior to analysis, which is not ideal for our need to analyze large numbers of samples. For example, in the method developed by Zhao et al. (24) for a suite of OPCs including TCPP in whole blood, after extracting the analytes twice from blood followed by concentration and reconstitution, samples underwent two extensive cleanup steps using silica and C18-SPE cartridges. Following the last cleanup, the filtrate was collected, evaporated to dryness and reconstituted in 100 μL of methanol for the analysis by UPLC–MS-MS. Although their LOD and LLOQ for TCPP were 0.19 and ~1.0 ng/mL, respectively, and recovery was ≥80%, the extensive sample cleanup and concentration steps involved make this method less suitable for application to large numbers of samples. Jonsson and Nilsson (39) used a stir bar-assisted microporous membrane liquid–liquid extraction (MMLLE), sample cleanup by SPE, concentration, reconstitution and analysis by GC-nitrogen phosphorus detection (NPD) to quantitate TCPP and seven other OPPs in human plasma. Their LOD for TCPP was 0.5 ng/g and recovery was 89%. While this method has similar LOD and recovery to our method, the extensive multistep cleanup, which requires MMLLE, creates a high threshold for its application in routine analyses of large sample numbers. Since NTP studies in rodents typically collect plasma from a large number of animals to evaluate internal dose, a method with minimal sample cleanup is needed. The method reported herein is simple, requiring only protein precipitation followed by extraction and concentration prior to analysis, making it more efficient when analyzing a large number of samples. The method recovery (≥86%), LLOQ (~5 ng/mL) and LOD (≤1 ng/mL) in plasma from rats and mice are comparable to previous methods and are sufficient for our needs. The method is versatile in that it has the following additional features; allows use of plasma volumes ranging from 50 to 200 μL, allows use of plasma collected using a variety of anticoagulants, samples with higher concentrations can be successfully diluted into the validated range for accurate quantitation and has acceptable lot-to-lot variability for most matrices. The higher variability (estimated as matrix factor) seen in the male mouse appears to result from the presence of background levels of TCPP in some of the samples. TCPP was frequently found in matrix and method blanks in our laboratory with the isomeric pattern confirming it as TCPP suggesting ubiquitous existence of this contaminant. In fact, TCPP has been found in the environment in airborne particulates (5, 6, 18), household dust (10, 11, 13, 19) and workplace settings (12). We conducted experiments to investigate potential sources of TCPP background in our laboratory. TCPP-1 concentrations in Kimwipes™ ranged from ~34 to 71 ng/m2 and in acetone were ~14 ng TCPP-1/mL, Wipe samples from the sample preparation laboratory had, on average, ~2,674 ng TCPP-1/m2 (range: ~21 to ~25,717 ng TCPP-1/m2). Dust is one possible source of the background TCPP levels found in our lab. Stapleton (9, 16) reported TCPP concentrations in household dust ranging from <140 to 5,490 ng/g (16) and 217–67,810 ng/g (9), respectively, while van den Eede et al. (43) reported TCPP dust levels of 0.19–73.7 μg/g. Because of the presence of background levels of TCPP in our laboratory, all laboratory glassware were cleaned before use with acetone (44, 45), which had previously been screened for TCPP. To ensure proper performance of the standard curve, a statistical analysis of the matrix standard curve for each sample batch was also compared to a historical average to detect potential contamination with TCPP prior to sample preparation. If the equation of the standard curve exceeded the 95% confidence interval for the historical average, the standards were rejected, and a new set was prepared and analyzed. Samples were only prepared and analyzed if the statistical analysis of the matrix standard curve passed criteria. Conclusion A simple, sensitive method for the determination of TCPP-1 in rodent plasma has been developed and validated over the range of 5–70 ng/mL. Plasma TCPP-1 concentrations up to 175 ng/mL could be successfully diluted into the validated analytical range. Recoveries were ≥86% for male and female rat and mouse plasma with and LLOQ of 5 ng/mL and LOD ≤ 1.0 ng/mL. TCPP-1 was stable in plasma when stored frozen for up to 60 days prior to analysis, demonstrating applicability of the method for the analysis of plasma samples following exposure of rodents to TCPP. Funding This research was supported by the NIH, National Institute of Environmental Health Sciences and DHHS contract number HHNS273201400020C with MRIGlobal, Kansas City, MO. Acknowledgments We also thank Dr Esra Mutlu and Dr Georgia Roberts for their review of this manuscript. References 1 (NRC) NRC . Toxicological Risks of Selected Flame-Retardant Chemicals . National Academies Press (US): Washington, DC , 2000 , pp. 338–341. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 2 van der Veen , I. , de Boer , J. ( 2012 ) Phosphorus flame retardants: properties, production, environmental occurrence, toxicity and analysis . Chemosphere , 88 , 1119 – 1153 . Google Scholar Crossref Search ADS PubMed WorldCat 3 (USEPA) USEPA . ( 2015 ) TSCA work plan chemical problem formulation and initial assessment: chlorinated phosphate ester cluster flame retardants. U.S. Environmental Protection Agency, Office of Pollution Prevention, Office of Pollution Prevention and Toxics, Washington, DC. 4 (USEPA) USEPA . ( 2012 ) Chemical Data Reporting (CDR): 2-propanol, 1-chloro, 2,2’,2”-phosphate. U.S. Environmental Protection Agency, Office of Chemical Safety and Pollution Protection, Office of Pollution Prevention and Toxics, Washington, DC. 5 Moeller , A. , Sturm , R., Xie , Z., Cai , M., He , J., Ebinghaus , R. ( 2012 ) Organophosphorus flame retardants and plasticizers in airborne particles over the northern Pacific and Indian Ocean toward the polar regions: evidence for global occurrence . Environmental Science and Technology , 46 , 3127 – 3134 . Google Scholar Crossref Search ADS PubMed WorldCat 6 Moeller , A , Xie , Z, Caba , A, Sturm , R, Ebinghaus , R. ( 2011 ) Organophosphorus flame retardants and plasticizers in the atmosphere of the North Sea . Environmental pollution (Barking, Essex: 1987) , 159 , 3660 – 3665 . Google Scholar Crossref Search ADS PubMed WorldCat 7 Brandsma , S.H. , Leonards , P.E.G., Leslie , H.A., de Boer , J. ( 2015 ) Tracing organophosphorus and brominated flame retardants and plasticizers in an estuarine food web . The Science of the Total Environment , 505 , 22 – 31 . Google Scholar Crossref Search ADS PubMed WorldCat 8 Brandsma , S.H. , de Boer , J., van Velzen , M.J.M., Leonards , PEG ( 2014 ) Organophosphorus flame retardants (PFRs) and plasticizers in house and car dust and the influence of electronic equipment . Chemosphere , 116 , 3 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 9 Stapleton , H.M. , Misenheimer , J., Hoffman , K., Webster , T.F. ( 2014 ) Flame retardant associations between children's handwipes and house dust . Chemosphere , 116 , 54 – 60 . Google Scholar Crossref Search ADS PubMed WorldCat 10 Langer , S. , Fredreicsson , M., Weschler , C.J., Bekö , G., Strandberg , B., Remberger , M., et al. . ( 2016 ) Organophosphate esters in dust samples collected from Danish homes and daycare centers . Chemosphere , 154 , 559 – 566 . Google Scholar Crossref Search ADS PubMed WorldCat 11 Luongo , G. , Ostman , C. ( 2016 ) Organophosphate and phthalate esters in settled dust from apartment buildings in Stockholm . Indoor Air , 26 , 414 – 425 . Google Scholar Crossref Search ADS PubMed WorldCat 12 Cristale , J. , Hurtado , A., Gomez-Canela , C., Lacorte , S. ( 2016 ) Occurrence and sources of brominated and organophosphorus flame retardants in dust from different indoor environments in Barcelona, Spain . Environmental Research , 149 , 66 – 76 . Google Scholar Crossref Search ADS PubMed WorldCat 13 Wu , M. , Yu , G., Cao , Z., Wu , D., Liu , K., Deng , S., et al. . ( 2016 ) Characterization and human exposure assessment of organophosphate flame retardants in indoor dust from several microenvironments of Beijing, China . Chemosphere , 150 , 465 – 471 . Google Scholar Crossref Search ADS PubMed WorldCat 14 He , R. , Li , Y., Xiang , P., Li , C., Zhou , C., Zhang , S., et al. . ( 2016 ) Organophosphorus flame retardants and phthalate esters in indoor dust from different microenvironments: bioaccessibility and risk assessment . Chemosphere , 150 , 528 – 535 . Google Scholar Crossref Search ADS PubMed WorldCat 15 Staaf , T. , Ostman , C. ( 2005 ) Organophosphate triesters in indoor environments . Journal of environmental monitoring: JEM , 7 , 883 – 887 . Google Scholar Crossref Search ADS PubMed WorldCat 16 Stapleton , H.M. , Klosterhaus , S., Eagle , S., Fuh , J., Meeker , J.D., Blum , A., et al. . ( 2009 ) Detection of organophosphate flame retardants in furniture foam and U.S. house dust . Environmental Science and Technology , 43 , 7490 – 7495 . Google Scholar Crossref Search ADS PubMed WorldCat 17 Stapleton , H.M. , Klosterhaus , S., Ferguson , P.L., van Bergen , S., Cooper , E., Webster , T.F., et al. . ( 2011 ) Identification of flame retardants in polyurethane foam collected from baby products . Environmental Science and Technology , 45 , 110518092340072 – 110518092345331 . Google Scholar Crossref Search ADS WorldCat 18 Fromme , H. , Kraft , M., Fembacher , L., Mach , C., Dietrich , S., Burkardt , R., et al. . ( 2014 ) Organophosphate flame retardants and plasticizers in the air and dust in German daycare centers and human biomonitoring in visiting children (LUPE 3) . Environment International , 71 , 158 – 163 . Google Scholar Crossref Search ADS PubMed WorldCat 19 Mizouchi , S. , Ichiba , M., Takigami , H., Kajiwara , N., Takamuku , T., Miyajima , T., et al. . ( 2015 ) Exposure assessment of organophosphorus and organobromine flame retardants via indoor dust from elementary schools and domestic houses . Chemosphere , 123 , 17 – 25 . Google Scholar Crossref Search ADS PubMed WorldCat 20 Hallanger , I.G. , Sagerup , K., Evenset , A., Kovacs , K.M., Leonards , P., Fuglei , E., et al. . ( 2015 ) Organophosphorous flame retardants in biota from Svalbard, Norway . Marine Pollution Bulletin , 101 , 442 – 447 . Google Scholar Crossref Search ADS PubMed WorldCat 21 Evenset , A.L.,.H. , Christensen , G.N., Warner , N., Remberger , M., Gabrielson , G.W. ( 2009 ) Screening of new contaminants in samples from Norwegian arctic. SPFO-Report 1049. 22 Leonards PSEHvdV , I. , Berg , V., Bustness , J.O., van Leeuwen , S. ( 2011 ) Screening of organophosphor flame retardants 2010. SPFO-Report 1091. 23 Sundkvist , A.M. , Olofsson , U., Haglund , P. ( 2010 ) Organophosphorus flame retardants and plasticizers in marine and fresh water biota and in human milk . Journal of environmental monitoring , 12 , 943 – 951 . Google Scholar Crossref Search ADS PubMed WorldCat 24 Zhao , F. , Wan , Y., Zhao , H., Wenxin , H., Mu , D., Webster , T.F., et al. . ( 2016 ) Levels of blood organophosphorus flame retardants and association with changes in human sphingolipid homeostasis . Environmental Science and Technology , 50 , 8896 – 8903 . Google Scholar Crossref Search ADS PubMed WorldCat 25 Su , G. , Letcher , R.J., Yu , H. ( 2015 ) Determination of organophosphate diesters in urine samples by a high-sensitivity method based on ultra high pressure liquid chromatography-triple quadrupole-mass spectrometry . Journal of Chromatography. A , 1426 , 154 – 160 . Google Scholar Crossref Search ADS PubMed WorldCat 26 Su , G. , Letcher , R.J., Yu , H., Gooden , D.M., Stapleton , H.M. ( 2016 ) Determination of glucuronide conjugates of hydroxyl triphenyl phosphate (OH-TPHP) metabolites in human urine and its use as a biomarker of TPHP exposure . Chemosphere , 149 , 314 – 319 . Google Scholar Crossref Search ADS PubMed WorldCat 27 WHO . ( 1998 ) Tris (chloropropyl) Phosphate and Tris (2-chloroethyl) Phosphate. Environmental Health Criteria, 209. 28 (OECD) OfEC-oaD . ( 2000 ) Screening Information Dataset (SIDS) initial assessment profile: Tris(1-chloro-2-propyl)phosphate, CAS No.: 13674-84-5. UNEP Chemicals, Geneva, Switzerland. 29 (EU) EU . ( 2008 ) Tris(2-chloro-1-methylethyl) phosphate (TCPP), CAS No. 13674-84-5, EINECS No. 237-158-7 Summary Risk Assessment Report, European Communities. 30 (ATSDR) AfTSaDR . ( 2012 ) Toxicological profile for phosphate ester flame retardants. U.S. Department of Health and Human Services, Agency for Toxic Substances and Disease Registry, Atlanta, GA. 31 Liu , X. , Ji , K., Choi , K. ( 2012 ) Endocrine disruption potentials of organophosphate flame retardants and related mechanisms in H295R and MVLN cell lines and in zebrafish . Aquatic Toxicology , 114 , 173 – 181 . Google Scholar Crossref Search ADS PubMed WorldCat 32 Powers , C.M. , Ryde , I.T., Roberts , S.C., Seidler , F.J., Slotkin , T.A., Stapleton , H.M. ( 2011 ) Is the PentaBDE replacement, tris (1,3-dichloropropyl) phosphate (TDCPP), a developmental neurotoxicant? Studies in PC12 cells . Toxicology and Applied Pharmacology , 256 , 281 – 289 . Google Scholar Crossref Search ADS PubMed WorldCat 33 Crump , D. , Chiu , S., Kennedy , S.W. ( 2012 ) Effects of tris(1,3-dichloro-2-propyl) phosphate and tris(1-chloropropyl) phosphate on cytotoxicity and mRNA expression in primary cultures of avian hepatocytes and neuronal cells . Toxicological sciences: an official journal of the Society of Toxicology , 126 , 140 – 148 . Google Scholar Crossref Search ADS PubMed WorldCat 34 Hu , M. , Li , J., Zhang , B., Cui , Q., Wei , S., Yu , H. ( 2014 ) Regional distribution of halogenated organophosphate flame retardants in seawater samples from three coastal cities in China . Marine Pollution Bulletin , 86 , 569 – 574 . Google Scholar Crossref Search ADS PubMed WorldCat 35 Guo , X. , Mu , T., Xian , Y., Luo , D., Wang , C. ( 2016 ) Ultra-performance liquid chromatography tandem mass spectrometry for the rapid simultaneous analysis of nine organophosphate esters in milk powder . Food Chemistry , 196 , 673 – 681 . Google Scholar Crossref Search ADS PubMed WorldCat 36 Qiao , L. , Zheng , X.B., Zheng , J., Lei , W.X., Li , H.F., Wang , M.H., et al. . ( 2016 ) Analysis of human hair to assess exposure to organophosphate flame retardants: influence of hair segments and gender differences . Environmental Research , 148 , 177 – 183 . Google Scholar Crossref Search ADS PubMed WorldCat 37 Schindler , B.K. , Förster , K., Angerer , J. ( 2009 ) Determination of human urinary organophosphate flame retardant metabolites by solid-phase extraction and gas chromatography-tandem mass spectrometry . Journal of Chromatography. B, Analytical Technologies in the Biomedical and Life Sciences , 877 , 375 – 381 . Google Scholar Crossref Search ADS PubMed WorldCat 38 Schindler , B.K. , Förster , K., Angerer , J. ( 2009 ) Quantification of two urinary metabolites of organophosphorus flame retardants by solid-phase extraction and gas chromatography–tandem mass spectrometry . Analytical and Bioanalytical Chemistry , 395 , 1167 – 1171 . Google Scholar Crossref Search ADS PubMed WorldCat 39 Jonsson , O.B. , Nilsson , U.L. ( 2003 ) Determination of organophosphate ester plasticisers in blood donor plasma using a new stir‐bar assisted microporous membrane liquid‐liquid extractor . Journal of Separation Science , 26 , 886 – 892 . Google Scholar Crossref Search ADS WorldCat 40 Carlsson , H. , Nilsson , U., Becker , G., Ostman , C. ( 1997 ) Organophosphate ester flame retardants and plasticizers in the indoor environment: analytical methodology and occurrence . Environmental Science & Technology , 31 , 2931 – 2936 . Google Scholar Crossref Search ADS WorldCat 41 Bergh , C , Torgrip , R, Emenius , G, Ostman , C. ( 2011 ) Organophosphate and phthalate esters in air and settled dust—a multi-location indoor study . Indoor Air , 21 , 67 – 76 . Google Scholar Crossref Search ADS PubMed WorldCat 42 Brody , S.S.C. , John , E. ( 1966 ) Flame photometric detector. Application of a specific detector for phosphorus and for sulfur compounds sensitive to subnanogram quantities . Journal of Gas Chromatography , 4 , 42 – 46 . Google Scholar Crossref Search ADS WorldCat 43 Van den Eede , N. , Dirtu , A.C., Neels , H., Covaci , A. ( 2011 ) Analytical developments and preliminary assessment of human exposure to organophosphate flame retardants from indoor dust . Environment International , 37 , 454 – 461 . Google Scholar Crossref Search ADS PubMed WorldCat 44 Carlsson , H. , Nilsson , U., Becker , G., Ostman , C. ( 1997 ) Organophosphate ester flame retardants and plasticizers in the indoor environment: analytical methodology and occurrence . Environmental Science and Technology , 31 , 2931 – 2936 . Google Scholar Crossref Search ADS WorldCat 45 Bergh , C. , Torgrip , R., Emenius , G., Ostman , C. ( 2011 ) Organophosphate and phthalate esters in air and settled dust—a multi-location indoor study . Indoor Air , 21 , 67 – 76 . Google Scholar Crossref Search ADS PubMed WorldCat Published by Oxford University Press 2018. This work is written by (a) US Government employee(s) and is in the public domain in the US. Published by Oxford University Press 2018.

Haddad,, Christina;Salman,, Rola;El-Hellani,, Ahmad;Talih,, Soha;Shihadeh,, Alan;Saliba, Najat, Aoun

doi: 10.1093/jat/bky065pmid: 30192935

Abstract Electronic cigarettes (ECIGs) are battery-powered devices that heat and vaporize solutions containing propylene glycol (PG) and/or vegetable glycerin (VG), nicotine and possible trace flavorants to produce an inhalable aerosol. The heating process can lead to the formation of reactive oxygen species (ROS), which are linked to various oxidative damage-initiated diseases. Several studies in the literature have addressed ROS emissions in ECIG aerosols, but the effects of power, ECIG device design and liquid composition on ROS are relatively unknown. In addition, ROS emissions have not been examined in the emerging high power, sub-Ohm device (SOD) category. In this study, an acellular 2',7'-dichlorofluorescin (DCFH) probe technique was optimized to measure ROS in ECIG aerosols. The technique was deployed to measure ROS emissions in SOD and supra-Ohm ECIGs while varying power, heater coil head design and liquid composition (PG/VG ratio and nicotine concentration). Liquids were made from analytical standards of PG, VG and nicotine and contained no flavorants. At high powers, ROS emissions in ECIGs and combustible cigarettes were similar. Across device designs, ROS emissions were uncorrelated with power (R2 = 0.261) but were highly correlated with power per unit area (R2 = 0.78). It was noticed that an increase in the VG percentage in the liquid yielded higher ROS flux, and nicotine did not affect ROS emissions. ROS emissions are a function of device design and liquid composition at a given power. For a given liquid composition, a promising metric for predicting ROS emissions across device designs and operating conditions is power per unit area of the heating coil. Importantly, ROS formation is significant even when the ECIG liquid consists of pure analytical solutions of PG and VG; it can therefore be viewed as intrinsic to ECIG operation and not solely a by-product of particular flavorants, contaminants or additives. Introduction Electronic cigarette (ECIG) use has become an epidemic worldwide, especially among youth (1, 2). ECIG use prevalence among cigarette smokers, former smokers and previously nicotine-naïve groups alike has increased tremendously in the last decade (3, 4). While it is often claimed that ECIGs are good smoking cessation tools (5), the issue is still controversial and empirical data to resolve it is sparse (6). On the other hand, ECIGs may renormalize smoking among users and bystanders (7, 8) and may initiate nicotine dependence among young users, potentially constituting a gateway to cigarette smoking (9–13). ECIGs are battery-powered devices that heat and vaporize solutions mainly consisting of propylene glycol (PG) and/or vegetable glycerin (VG) and nicotine to generate inhalable aerosols (14) that are not toxicant-free (15). Toxicants detected in ECIG aerosols are either present in the liquid solutions even prior to heating (16, 17) or are produced via the thermal decomposition of the liquid constituents on the hot surface of the heating coil (18–20). The most studied toxicants formed in situ are carbonyl compounds that result from the dehydration and oxidation of the alcohol functional groups on a metal surface (21). Other toxicants include furanic and aromatic compounds, which have been identified when additives such as sugar and fruit flavors are present in the mix (22, 23). In addition to carbonyls, thermal breakdown of chemical bonds in ECIG liquids may lead to the formation of reactive oxygen species (ROS), a class of chemicals, which induce oxidative stress in cells. It has been well established that oxidative stress from cigarette smoke exposure leads to pulmonary diseases (24–27). A growing number of studies have linked ROS emissions from ECIG to cytotoxicity in pulmonary tissues (28–30). Several studies in the literature have reported ROS emissions in ECIG aerosols using cellular and acellular assays (31–36). In addition, electron paramagnetic resonance (EPR) studies have revealed the presence of radical species in ECIG aerosols (37, 38), and a recent report by Zhao et al. assessed the effect of various parameters, including brand, flavor, power and users’ puffing regimens, on the generation of ROS (38). In this work, we used an optimized 2′,7′-dichlorofluorescin (DCFH) probe solution in order to measure ROS emissions from conventional tank and sub-Ohm ECIG devices (SODs) as a function of power, coil head geometry and ECIG liquid composition (PG/VG ratio, nicotine content). ROS emissions were compared across conditions to ROS emissions from combustible cigarette and plotted vs power per coil surface area, which we have recently shown as the relevant predictor of ECIG toxicant emissions that are formed in situ (39). Materials and Methods Materials PG (99.5%) (CAS № 57-55-6), VG (99–101%) (CAS № 56-81-5), ethanol and deionized (DI) water were procured from Sigma-Aldrich. Pure nicotine (CAS № 54-11-5), horseradish peroxidase (HRP) (52 units/mg) (CAS № 9003-99-0), potassium phosphate monobasic (CAS № 7778-77-0) and dibasic (CAS № 7758-11-4) were purchased from Sigma-Aldrich. 2′,7′-dichlorofluorescin diacetate (DCFH-DA) was purchased from Molecular Probes (product code D399). Quartz filters (ADVENTEC, QR-100.47 mm) were procured from Whatman International. Preparation of DCFH probe solution DCFH-DA was dissolved in ethanol in order to prepare a 125 μM solution. The DCFH-DA solution (10 mL) was deacetylated with 40 ml of 0.01 M NaOH aqueous solution. The activated DCFH solution was wrapped in aluminum foil and kept in the dark for 30 min. A phosphate buffer (pH = 7.1), prepared by mixing monobasic and dibasic potassium phosphate to attain a 0.25 mM concentration (200 ml), was added to 50 mL of DCFH solution. Horseradish peroxidase (0.5 units/mL) was added (2.4 mg) to amplify the fluorescence signal. The final 250 mL working solution had a concentration of 5 μM of DCFH. A linear calibration curve (1 × 10−7 to 10−6 M) was constructed using hydrogen peroxide (H2O2) to express ROS equivalents. The limit of detection (LOD) was 0.14 × 10−7 M, and the limit of quantification (LOQ) was 0.48 × 10−7 M of H2O2. LOD and LOQ of the method were calculated according to the equations mentioned in the eighth edition of Quantitative Chemical Analysis by Daniel C. Harris (p. 103–106), which is LOD = 3 SD/m and LOQ = 10 SD/m, where SD stands for standard deviation of replicates of a low concentration (C = 0.5 × 10−7 M of H2O2) and m is the slope of the calibration curve. Probe solution The optimal experimental conditions of the DCFH solution were determined so that the photo- and auto-oxidation of the probe solution were minimized (40). Several combinations of DCFH concentration, storage temperature and duration, and mixing time were tested in order to achieve this goal. The final probe was a 5 μM DCFH solution, stored at 4°C and mixed for 30 min with the samples (41, 42). This solution provided a >98% calibration R2 with <6% bias error due to auto-oxidation at the maximum allowed solution storage time of 2.5 h. All storage and reacting samples were wrapped in an aluminum foil to prevent photo-oxidation. Aerosol generation The American University of Beirut’s aerosol lab vaping instrument (ALVIN) (43) was used to generate ECIG aerosols. Puff duration, inter-puff interval and flow rate were selected to represent the pattern of an “experienced” ECIG user (4-s puff duration, 10-s inter-puff interval, 1 L/min flow rate) (43). A vaping session constituted from five puffs on a supra-Ohm ECIG and two puffs on a sub-Ohm ECIG, both sessions having a 4-s puff duration, a 10-s puff interval and a volume of 67 mL/puff. In the case of the conventional cigarette, 10 puffs were executed using ISO protocol puffing parameters (2-s puff duration, 60-s inter-puff interval, 35 mL/puff). For both the ECIG and combustible cigarette conditions, the aerosol was drawn through a particulate filter trap as described by Zhao and Hopke (41). Study design and sampling ROS emissions in the total particulate matter (TPM) of ECIG aerosols and the smoke of the tobacco cigarettes were assessed, as it was previously found that ROS concentrations in the particle phase are much greater than in the gas phase (41). TPM was trapped on a 47-mm quartz filter installed at the mouth end of the ECIG and the tobacco cigarettes and then directly immersed in 20 mL of a freshly prepared DCFH probe solution. Fluorescence was read on a SpectraMax M5 microplate reader acting as a fluorimeter. Liquids containing 50/50 PG/VG solution with 12 mg/mL of nicotine were vaped in a VaporFi platinum tank at two different powers (5 and 11 W) and in a sub-Ohm SmokTFV8 device equipped with a V8-T8 coil head (eight coils) at five different powers (50, 75, 100, 150 and 200 W). Keeping the power and liquid constant at 50 W and using a 50/50 PG/VG solution with 12 mg/mL of nicotine, the effect of different coil heads was assessed using the sub-Ohm device (SOD) equipped with V8-Q4 (4 coils), V8-T8 (8 coils), V8-T10 (10 coils) and TF-Q4 (4 coils) (Figure 1). Three PG/VG ratios were prepared from standard liquid PG and VG (100/0, 50/50 and 0/100 PG/VG ratios), and three different nicotine loads in a 50/50 PG/VG solution were tested (0, 6 and 12 mg/mL nicotine concentrations). These solutions were vaped at two different powers for each device (5 and 11 W for supra-Ohm device and 50 and 150 W for SOD). Each condition was repeated in triplicate, and the results are reported as the mean of three measurements after blank subtraction. Figure 1. View largeDownload slide The sub-Ohm SmokTFV8 device with illustration of the different coil heads (retrieved from the study of Talih et al. (39)). Figure 1. View largeDownload slide The sub-Ohm SmokTFV8 device with illustration of the different coil heads (retrieved from the study of Talih et al. (39)). TPM, surface area, and ROS flux The amount of TPM was determined gravimetrically by weighing the filter pad and its holder before and after each sampling session. The total surface area of the coil was calculated based on the coil wire diameter (measured using calipers), the length of coil wire and the number of coils (39). ROS emissions are reported as the number of moles of H2O2 equivalent per second of vaping/smoking in order to facilitate comparison between different puffing regimens. Statistical analysis T-test was used to estimate the statistical significance of the difference between powers relative to the lowest power for each ECIG and relative to the combustible cigarette level. It was also used to assess the effect of liquid composition (PG/VG ratio and nicotine content) on ROS emission. Results Effect of power and power per unit coil surface area ROS emission rates as a function of power are shown in Table I for VaporFi Platinum and SmokTFV8 SOD. In the supra-Ohm device, the ROS flux in the aerosols generated using 11 W was three times higher than that of 5 W (P < 0.1). In the SOD, the ROS flux showed an increase between 50 and 200 W (P < 0.1). ROS emissions at the highest power tested (200 W) in SOD device was comparable to those of conventional cigarettes (Table I). Table I. ROS flux as a function of power and coil head in VaporFi Platinum and SmokTFV8 SOD devices in comparison to a conventional cigarette. Statistical significance is shown in comparison with the conventional cigarette ECIG Coil head Power (W) ROS flux (nmole/s) VaporFi Single coil 5 0.238 ± 0.253** 11 0.696 ± 0.096* SmokTFV8 SOD V8-T8 50 0.114 ± 0.034** 75 0.109 ± 0.042** 100 0.167 ± 0.117** 150 0.241 ± 0.029* 200 1.143 ± 0.606 V8-Q4 50 0.066 ± 0.030** V8-T10 50 0.045 ± 0.019** TF-Q4 50 0.049 ± 0.016** Tobacco cig 1.240 ± 0.210 ECIG Coil head Power (W) ROS flux (nmole/s) VaporFi Single coil 5 0.238 ± 0.253** 11 0.696 ± 0.096* SmokTFV8 SOD V8-T8 50 0.114 ± 0.034** 75 0.109 ± 0.042** 100 0.167 ± 0.117** 150 0.241 ± 0.029* 200 1.143 ± 0.606 V8-Q4 50 0.066 ± 0.030** V8-T10 50 0.045 ± 0.019** TF-Q4 50 0.049 ± 0.016** Tobacco cig 1.240 ± 0.210 Significant difference from tobacco cigarette level: *P < 0.05, **P < 0.01. Table I. ROS flux as a function of power and coil head in VaporFi Platinum and SmokTFV8 SOD devices in comparison to a conventional cigarette. Statistical significance is shown in comparison with the conventional cigarette ECIG Coil head Power (W) ROS flux (nmole/s) VaporFi Single coil 5 0.238 ± 0.253** 11 0.696 ± 0.096* SmokTFV8 SOD V8-T8 50 0.114 ± 0.034** 75 0.109 ± 0.042** 100 0.167 ± 0.117** 150 0.241 ± 0.029* 200 1.143 ± 0.606 V8-Q4 50 0.066 ± 0.030** V8-T10 50 0.045 ± 0.019** TF-Q4 50 0.049 ± 0.016** Tobacco cig 1.240 ± 0.210 ECIG Coil head Power (W) ROS flux (nmole/s) VaporFi Single coil 5 0.238 ± 0.253** 11 0.696 ± 0.096* SmokTFV8 SOD V8-T8 50 0.114 ± 0.034** 75 0.109 ± 0.042** 100 0.167 ± 0.117** 150 0.241 ± 0.029* 200 1.143 ± 0.606 V8-Q4 50 0.066 ± 0.030** V8-T10 50 0.045 ± 0.019** TF-Q4 50 0.049 ± 0.016** Tobacco cig 1.240 ± 0.210 Significant difference from tobacco cigarette level: *P < 0.05, **P < 0.01. Holding power and liquid composition constant, there was a significant difference only between the coil head V8-T8 and both V8-T10 and TF-Q4 (P < 0.1). ROS emission was weakly correlated with power across devices (R2 = 0.26); however, a significant correlation was found when the ROS flux was plotted as a function of power per surface area, P/SA (R2 = 0.78), as shown in Figure 2. The two data points of the supra-Ohm device seem to fit well within this correlation. Figure 2. View largeDownload slide ROS flux as a function of power per surface area of the coil. Filled circles correspond to SOD, while empty circles correspond to supra-Ohm. Figure 2. View largeDownload slide ROS flux as a function of power per surface area of the coil. Filled circles correspond to SOD, while empty circles correspond to supra-Ohm. Effect of liquid composition The effect of the PG/VG ratio on ROS emissions from ECIG is shown in Figure 3A and B for the VaporFi and the SmokTFV8 SOD. In both devices, ROS flux trended downward with increasing VG content and attained significant difference between pure PG and VG liquids (P < 0.05). On the other hand, nicotine concentration did not have any effect on ROS emissions (Figures 3C and D). Figure 3. View largeDownload slide ROS flux as a function of the PG/VG ratio in the liquid vaped on the supra-Ohm device (A) and SOD (B). ROS flux vs nicotine content in the vaped liquid on the supra-Ohm device (C) and SOD (D). Figure 3. View largeDownload slide ROS flux as a function of the PG/VG ratio in the liquid vaped on the supra-Ohm device (A) and SOD (B). ROS flux vs nicotine content in the vaped liquid on the supra-Ohm device (C) and SOD (D). Discussion Our results showed that ROS flux in tank and SOD ECIGs increases with power within the same device design. At high powers, ROS emissions from both ECIGs, especially the SOD, can reach levels that are similar to those of tobacco cigarettes. Higher powers have been associated with elevated temperatures on the coil surface causing an increase in the TPM emitted and/or a higher probability of the degradation of the chemical bonds in the molecules of the vaped liquids (38). ROS emission is not always significantly affected by the coil head design. In this study, we showed that the ROS flux is significantly correlated (78%) with P/SA, supporting the theory that P/SA is a better predictor of toxicant emissions in general than the power or the number of coils (39). The surge of the ROS flux level at 200 W is not linked to an increase in TPM (Figure 4), and therefore high ROS flux in this particular case can be ascribed to a spike in temperature caused by the “dry puff” phenomenon. Figure 4. View largeDownload slide ROS flux (nmole/s) vs TPM flux (mg/s) in both devices at different powers and with different coil heads in the case of the SOD. Figure 4. View largeDownload slide ROS flux (nmole/s) vs TPM flux (mg/s) in both devices at different powers and with different coil heads in the case of the SOD. The chemical degradation of the ECIG liquids—PG, VG and nicotine—are thought to play a determinant role in ROS emissions. In this study, we showed that an increase in the VG percentage in the liquid yielded higher ROS flux, and this may be due to the emission of higher ROS from VG molecules or to a slower wicking and consequently higher probabilities of the “dry puff” phenomenon, particularly at high powers. This is in disagreement with a recent paper in which a higher PG/VG ratio in the liquid was correlated with higher radical emissions from ECIGs (44). Our study also showed that nicotine does not affect ROS emissions, which are mainly a function of the chemical nature of the solvent and the P/SA of the coil. Conclusion Our results showed that ECIGs intrinsically emit ROS even in the absence of flavorants. ROS levels from conventional tank ECIGs and SODs at high powers could reach tobacco cigarette-like levels. P/SA is a better predictor of ROS emissions than power. In addition, ROS emission is affected by the chemical constituents of the vaped liquid (PG/VG ratio). Toxicant flux is an easy tool to compare the results across different puffing regimens and among studies. Funding This work was supported by the National Institute on Drug Abuse of the National Institutes of Health (grant number P50DA036105) and the Center for Tobacco Products of the US Food and Drug Administration. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Food and Drug Administration. Conflict of Interest The authors have no conflict of interest to declare. References 1 US-Surgeon-General . E-cigarette Use Among Youth and Young Adults: A Report of the Surgeon General 2016. https://e-cigarettes.surgeongeneral.gov/getthefacts.html; https://e-cigarettes.surgeongeneral.gov/documents/2016_SGR_Full_Report_508.pdf (accessed Nov 14, 2017). 2 Rennie , L.J. , Bazillier-Bruneau , C. , Rouëssé , J. ( 2016 ) Harm reduction or harm introduction? Prevalence and correlates of e-cigarette use among French adolescents . Journal of Adolescent Health , 58 , 440 – 445 . Google Scholar Crossref Search ADS PubMed 3 Cho , H.-J. , Dutra , L.M. , Glantz , S.A. ( 2017 ) Differences in adolescent e-cigarette and cigarette prevalence in two policy environments: South Korea and the United States . Nicotine and Tobacco Research , ntx198 . https://doi.org/110.1093/ntr/ntx1198 . 4 Dai , H. , Hao , J. ( 2017 ) Direct marketing promotion and electronic cigarette use among US adults, National Adult Tobacco Survey, 2013–2014 . Preventing Chronic Disease , 14 , 170073 . Google Scholar Crossref Search ADS 5 Giovenco , D.P. , Delnevo , C.D. ( 2018 ) Prevalence of population smoking cessation by electronic cigarette use status in a national sample of recent smokers . Addictive Behaviors , 76 , 129 – 134 . Google Scholar Crossref Search ADS PubMed 6 Hartmann-Boyce , J. , McRobbie , H. , Bullen , C. , Begh , R. , Stead , L.F. , Hajek , P. ( 2016 ) Electronic cigarettes for smoking cessation . Cochrane Database of Systematic Reviews , 9 , CD010216 . Google Scholar PubMed 7 Cataldo , J.K. , Petersen , A.B. , Hunter , M. , Wang , J. , Sheon , N. ( 2015 ) E-cigarette marketing and older smokers: road to renormalization . American Journal of Health Behavior , 39 , 361 – 371 . Google Scholar Crossref Search ADS PubMed 8 Sæbø , G. , Scheffels , J. ( 2017 ) Assessing notions of denormalization and renormalization of smoking in light of e-cigarette regulation . International Journal of Drug Policy , 49 , 58 – 64 . Google Scholar Crossref Search ADS PubMed 9 Etter , J.-F. ( 2017 ) Gateway effects and electronic cigarettes . Addiction . DOI:10.1111/add.13924 n/a-n/a. 10 Akre , C. , Suris , J.-C. ( 2017 ) Adolescents and young adults’ perceptions of electronic cigarettes as a gateway to smoking: a qualitative study in Switzerland . Health Education Research , 32 , 448 – 454 . Google Scholar Crossref Search ADS PubMed 11 Barrington-Trimis , J.L. , Urman , R. , Berhane , K. , Unger , J.B. , Cruz , T.B. , Pentz , M.A. , et al. . ( 2016 ) E-cigarettes and future cigarette use . Pediatrics , 138 , e20160379 . Google Scholar Crossref Search ADS PubMed 12 Stanbrook , M.B. ( 2016 ) Electronic cigarettes and youth: a gateway that must be shut . Canadian Medical Association Journal , 188 , 785 . Google Scholar Crossref Search ADS PubMed 13 Gautier , S. , Kinouani , S. , Raherison , C. ( 2017 ) Do electronic cigarettes increase the risk of smoking among adolescents and young adults? Santé Publique , 29 , 333 – 340 . Google Scholar Crossref Search ADS PubMed 14 Breland , A. , Soule , E. , Lopez , A. , Ramôa , C. , El-Hellani , A. , Eissenberg , T. ( 2017 ) Electronic cigarettes: what are they and what do they do? Annals of the New York Academy of Sciences , 1394 , 5 – 30 . Google Scholar Crossref Search ADS PubMed 15 Goniewicz , M.L. , Gawron , M. , Smith , D.M. , Peng , M. , Jacob , I.I.I.P. , Benowitz , N.L. ( 2017 ) Exposure to nicotine and selected toxicants in cigarette smokers who switched to electronic cigarettes: a longitudinal within-subjects observational study . Nicotine and Tobacco Research , 19 , 160 – 167 . Google Scholar Crossref Search ADS PubMed 16 Beauval , N. , Howsam , M. , Antherieu , S. , Allorge , D. , Soyez , M. , Garçon , G. , et al. . ( 2016 ) Trace elements in e-liquids—development and validation of an ICP-MS method for the analysis of electronic cigarette refills . Regulatory Toxicology and Pharmacology , 79 , 144 – 148 . Google Scholar Crossref Search ADS PubMed 17 Lisko , J.G. , Tran , H. , Stanfill , S.B. , Blount , B.C. , Watson , C.H. ( 2015 ) Chemical composition and evaluation of nicotine, tobacco alkaloids, pH, and selected flavors in e-cigarette cartridges and refill solutions . Nicotine and Tobacco Research , 17 , 1270 – 1278 . Google Scholar Crossref Search ADS PubMed 18 Khlystov , A. , Samburova , V. ( 2016 ) Flavoring compounds dominate toxic aldehyde production during e-cigarette vaping . Environmental Science and Technology , 50 , 13080 – 13085 . Google Scholar Crossref Search ADS PubMed 19 Margham , J. , McAdam , K. , Forster , M. , Liu , C. , Wright , C. , Mariner , D. , et al. . ( 2016 ) Chemical composition of aerosol from an e-cigarette: a quantitative comparison with cigarette smoke . Chemical Research in Toxicology , 29 , 1662 – 1678 . Google Scholar Crossref Search ADS PubMed 20 Talih , S. , Balhas , Z. , Salman , R. , Karaoghlanian , N. , Shihadeh , A. ( 2016 ) “Direct dripping”: a high-temperature, high-formaldehyde emission electronic cigarette use method . Nicotine and Tobacco Research , 18 , 453 – 459 . Google Scholar Crossref Search ADS PubMed 21 Uchiyama , S. , Senoo , Y. , Hayashida , H. , Inaba , Y. , Nakagome , H. , Kunugita , N. ( 2016 ) Determination of chemical compounds generated from second-generation e-cigarettes using a sorbent cartridge followed by a two-step elution method . Analytical Sciences , 32 , 549 – 555 . Google Scholar Crossref Search ADS PubMed 22 Soussy , S. , EL-Hellani , A. , Baalbaki , R. , Salman , R. , Shihadeh , A. , Saliba , N.A. ( 2016 ) Detection of 5-hydroxymethylfurfural and furfural in the aerosol of electronic cigarettes . Tobacco Control , 25 , ii88 – ii93 . Google Scholar Crossref Search ADS PubMed 23 Pankow , J.F. , Kim , K. , McWhirter , K.J. , Luo , W. , Escobedo , J.O. , Strongin , R.M. , et al. . ( 2017 ) Benzene formation in electronic cigarettes . PLoS One , 12 , e0173055 . Google Scholar Crossref Search ADS PubMed 24 Bartal , M. ( 2005 ) COPD and tobacco smoke . Monaldi Archives for Chest Disease , 63 , 213 – 225 . Google Scholar Crossref Search ADS PubMed 25 Bialas , A.J. , Sitarek , P. , Mikowska-Dymanowska , J. , Piotrowski , W.J. , Gorski , P. ( 2016 ) The role of mitochondria and oxidative/antioxidative imbalance in pathobiology of chronic obstructive pulmonary disease . Oxidative Medicine and Cellular Longevity , doi: 10.1155/2016/7808576. 26 US-Surgeon-General . How Tobacco Smoke Causes Disease: The Biology and Behavioral Basis for Smoking-Attributable Disease: A Report of the Surgeon General. 2010. https://www.ncbi.nlm.nih.gov/books/NBK53021/#ch7.s17 (accessed Nov 14, 2017). 27 Sundar , I.K. , Yao , H. , Rahman , I. ( 2012 ) Oxidative stress and chromatin remodeling in chronic obstructive pulmonary disease and smoking-related diseases . Antioxidants and Redox Signaling , 18 , 1956 – 1971 . Google Scholar Crossref Search ADS PubMed 28 Moses , E. , Wang , T. , Corbett , S. , Jackson , G.R. , Drizik , E. , Perdomo , C. , et al. . ( 2017 ) Molecular impact of electronic cigarette aerosol exposure in human bronchial epithelium . Toxicological Sciences , 155 , 248 – 257 . Google Scholar Crossref Search ADS PubMed 29 Javed , F. , Kellesarian , S.V. , Sundar , I.K. , Romanos , G.E. , Rahman , I. ( 2017 ) Recent updates on electronic cigarette aerosol and inhaled nicotine effects on periodontal and pulmonary tissues . Oral Diseases , 23 , 1052 – 1057 . Google Scholar Crossref Search ADS PubMed 30 Ganapathy , V. , Manyanga , J. , Brame , L. , McGuire , D. , Sadhasivam , B. , Floyd , E. , et al. . ( 2017 ) Electronic cigarette aerosols suppress cellular antioxidant defenses and induce significant oxidative DNA damage . PLoS One , 12 , e0177780 . Google Scholar Crossref Search ADS PubMed 31 Lerner , C.A. , Rutagarama , P. , Ahmad , T. , Sundar , I.K. , Elder , A. , Rahman , I. ( 2016 ) Electronic cigarette aerosols and copper nanoparticles induce mitochondrial stress and promote DNA fragmentation in lung fibroblasts . Biochemical and Biophysical Research Communications , 477 , 620 – 625 . Google Scholar Crossref Search ADS PubMed 32 Lerner , C.A. , Sundar , I.K. , Watson , R.M. , Elder , A. , Jones , R. , Done , D. , et al. . ( 2015 ) Environmental health hazards of e-cigarettes and their components: Oxidants and copper in e-cigarette aerosols . Environmental Pollution , 198 , 100 – 107 . Google Scholar Crossref Search ADS PubMed 33 Lerner , C.A. , Sundar , I.K. , Yao , H. , Gerloff , J. , Ossip , D.J. , McIntosh , S. , et al. . ( 2015 ) Vapors produced by electronic cigarettes and e-juices with flavorings induce toxicity, oxidative stress, and inflammatory response in lung epithelial cells and in mouse lung . PLoS One , 10 , e0116732 . Google Scholar Crossref Search ADS PubMed 34 Rubenstein , D.A. , Hom , S. , Ghebrehiwet , B. , Yin , W. ( 2015 ) Tobacco and e-cigarette products initiate Kupffer cell inflammatory responses . Molecular Immunology , 67 , 652 – 660 . Google Scholar Crossref Search ADS PubMed 35 Shaito , A. , Saliba , J. , Husari , A. , El-Harakeh , M. , Chhouri , H. , Hashem , Y. , et al. . ( 2017 ) Electronic cigarette smoke impairs normal mesenchymal stem cell differentiation . Scientific Reports , 7 , 14281 . Google Scholar Crossref Search ADS PubMed 36 Anderson , C. , Majeste , A. , Hanus , J. , Wang , S. ( 2016 ) E-cigarette aerosol exposure induces reactive oxygen species, DNA damage, and cell death in vascular endothelial cells . Toxicology Science , 154 , 332 – 340 . Google Scholar Crossref Search ADS 37 Goel , R. , Durand , E. , Trushin , N. , Prokopczyk , B. , Foulds , J. , Elias , R.J. , et al. . ( 2015 ) Highly reactive free radicals in electronic cigarette aerosols . Chemical Research in Toxicology , 28 , 1675 – 1677 . Google Scholar Crossref Search ADS PubMed 38 Zhao , J. , Zhang , Y. , Sisler , J.D. , Shaffer , J. , Leonard , S.S. , Morris , A.M. , et al. . ( 2018 ) Assessment of reactive oxygen species generated by electronic cigarettes using acellular and cellular approaches . Journal of Hazardous Materials , 344 , 549 – 557 . Google Scholar Crossref Search ADS PubMed 39 Talih , S. , Salman , R. , Karaoghlanian , N. , El-Hellani , A. , Saliba , N. , Eissenberg , T. , et al. . ( 2017 ) “Juice monsters”: sub-Ohm vaping and toxic volatile aldehyde emissions . Chemical Research in Toxicology , 30 , 1791 – 1793 . Google Scholar Crossref Search ADS PubMed 40 Chen , X. , Zhong , Z. , Xu , Z. , Chen , L. , Wang , Y. ( 2010 ) 2′,7′-Dichlorodihydrofluorescein as a fluorescent probe for reactive oxygen species measurement: forty years of application and controversy . Free Radical Research , 44 , 587 – 604 . Google Scholar Crossref Search ADS PubMed 41 Zhao , J. , Hopke , P.K. ( 2012 ) Concentration of reactive oxygen species (ROS) in mainstream and sidestream cigarette smoke . Aerosol Science and Technology , 46 , 191 – 197 . Google Scholar Crossref Search ADS 42 Wang , Y. , Hopke , P.K. , Sun , L. , Chalupa , D.C. , Utell , M.J. ( 2011 ) Laboratory and field testing of an automated atmospheric particle-bound reactive oxygen species sampling-analysis system . Journal of Toxicology , 2011 , 419476 . Google Scholar Crossref Search ADS PubMed 43 Talih , S. , Balhas , Z. , Eissenberg , T. , Salman , R. , Karaoghlanian , N. , El Hellani , A. , et al. . ( 2015 ) Effects of user puff topography, device voltage, and liquid nicotine concentration on electronic cigarette nicotine yield: measurements and model predictions . Nicotine and Tobacco Research , 17 , 150 – 157 . Google Scholar Crossref Search ADS PubMed 44 Bitzer , Z.T. , Goel , R. , Reilly , S.M. , Foulds , J. , Muscat , J. , Elias , R.J. , et al. . ( 2018 ) Effects of solvent and temperature on free radical formation in electronic cigarette aerosols . Chemical Research in Toxicology , 31 , 4 – 12 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Lee,, Wonho;Ko, Beom, Jun;Sim, Yeong, eun;Suh,, Sungill;Yoon,, Dahye;Kim,, Suhkmann

doi: 10.1093/jat/bky061pmid: 30165495

Abstract Urine was most commonly used biological sample in drug test. To create a false-negative test result, some drug abusers were reported to submit animal urine instead of their own. So, the purpose of this study was to compare and differentiate human from animal urine (Rat 370, Pig 12, Horse 10, Cat 8, Dog 13, Cow 10, Monkey 10) samples through the uses of quantitative 1H-NMR and to find biomarkers that can be used for the discrimination of human urine from animal urine. The 1H-NMR spectroscopy was employed and metabolomic analysis with multivariate statistics was carried out. Human urine samples and animal urine samples showed different patterns in metabolites profile and several characteristic metabolites were identified. Introduction Urine is the most commonly used biological sample in drug test, which has such advantages as easier collection, larger quantity, lower protein amounts and higher amounts of hydrophilic metabolites both from Phase I and Phase II reactions (1), when compared with other biological samples. Thus, urine drug test protocols have been firmly developed and used as the prime drug test methods (2–4). But recently a lot of cases of cheating drug test through urine sample tampering, are reported. Urine sample tempering includes urine dilution, urine adulteration and substitution of urine sample with animal urine or other similar looking liquids. Generally, creatinine concentration, acidity (pH) and specific gravity are measured for validity of urine samples (4–6), and mandatory guidelines for collecting and checking urine have been established in the USA (7). However, not only the procedures with these guidelines require more cost and time but also urine samples from animals such as cats, dogs and horses, have similar creatinine and pH values to human’s (7–10). Therefore, it is important to establish a simple, direct and accuracy method. The NMR metabolomics needs very simple sample preparation process and short run-time (11–13), which can make it possible to make a useful quick screening method. And 1H- NMR spectroscopy with multivariate statistics was well suited for complex mixture analysis (13, 14). In this study, the urinary analysis for discriminating human urine from animal urine was developed using 1H-NMR based metabolomics. 1H-NMR analyses were carried out for all samples (Human 359, Rat 370, Pig 12, Horse 10, Cat 8, Dog 13, Cow 10, Monkey 10). Acquired NMR data of the whole set were submitted to pattern recognition methods such as unsupervised principal component analysis and supervised orthogonal partial least squares discriminant analysis (OPLS-DA). Then, targeted metabolite profiling method was applied for the acquired NMR spectra (15). Methods Materials Chemicals D2O and sodium 3-(trimethylsilyl)-2,2,3,3-d4-propionate (TSP-d4) was purchased from Sigma-Aldrich (St. Louis, MO, USA). Urine samples Three hundred and fifty-nine urine samples collected from suspected drug abusers were obtained from prosecution service and police agency in Busan metropolitan city between January and June 2014. Three hundred and seventy rat samples were obtained from the Department of Chemistry, Pusan National University. Twelve Pig, 10 Horse, 8 Cat, 13 Dog and 10 Cow samples were obtained from the National Institute of Animal Science, RDA, Korea. Ten monkey samples were obtained from Korea Institute of Toxicology. Random samples of animal urine were taken from the animals that were placed in individual metabolic chambers. Their specifically formulated diets were provided usually twice a day, early in the morning and in the evening. Formulation of diets may vary depending on the experiments involved, age, sex and physiological conditions (pregnancy, milking, fattening, disease et cetera). Horses and cows are given free choice of grass hay. The samples were stored at −80°C for 5 days to a week and the day before analysis, frozen urine samples were stored at 4°C. Sample preparation All urine samples were centrifuged at 13,000 rpm for 5 min. The 630-μL of urine supernatant was mixed with 70 μL of D2O containing 20 mM TSP-d4(3-(trimethylsilyl)-propionic-2,2,3,3-d4 acid sodium salt) used as a reference for the chemical shift (0.00 ppm) and for quantification. All samples were stored at 4°C and then transferred into 5-mm NMR tube prior to NMR measurement. NMR spectroscopy The NMR experiments were carried out on a 600-MHz spectrometer (Agilent Technologies, Inc, CA) operating at 600.167 MHz (14.1 T) and equipped with 7600-AS auto sampler at 299.1 K. The spectra were acquired with a presaturation pulse sequence to suppress the water peak. The 1H-NMR spectra were measured using a 1.52 s relaxation delay, 1.998 sec acquisition time and 8 min total acquisition time. A total of 128 scans were acquired for each sample at a spectral width of 9615.4 Hz. D2O provided for a field frequency lock. Statistical analysis and metabolite identification The TSP-d4 peak at 0.0 ppm was used as a reference to calibrate the chemical shifts. Each 1H-NMR spectrum was binned from 0.5 to 10 ppm. And because of the difference in concentration between samples, the binning data were normalized to the total area. The multivariate analyses were performed for mean-centered data by using SIMCA-P + 12.0 software (Umetrics, Sweden). The qualities of the models were demonstrated by the cross-validation parameters R2X and Q2 values. R2 is defined as the proportion of variance in the data explained by the models and indicates the goodness of fit. Q2 is defined as the proportion of variance in the data predictable by the model and indicates predictability (16). In order to maximize the separation between groups, OPLS-DA was performed (17). Chenomx NMR suite 7.1 software (Chenomx Inc., Canada) was used a qualitative and quantitative analysis of urine components. The software reference libraries comprise hundreds of fully searchable pH dependent compound models. Single spectral areas were confirmed by analysis of each spike, and overlapping spectral areas were analyzed by a 2D correlation spectroscopy (COSY) NMR spectrum. Results and discussion 1H-NMR spectral data were analyzed for all urine sample, which include Human 359 urine samples, Rat 370 urine samples, Pig 12 urine samples, Horse 10 urine samples, Cat 8 urine samples, Dog 13 urine samples, Cow 10 urine samples and Monkey 10 urine samples. And, acquired NMR data of the whole set were submitted to multivariate statistical analysis of supervised OPLS-DA. Then, targeted metabolite profiling method was applied for the acquired NMR spectra. The spectral patterns of the human group and animal groups were compared by multivariate statistical analysis. Multivariate methods use redundancy in data to reduce dimensionality by separating regularities from noise. Dimensionality reduction is generating new variables of principal or predictive components. So, summarizes the information in the observations as a few new (latent) variables (X and Y axes are the scores, t, are new variables that best summarize the old ones; linear combinations of the old ones with coefficients.). OPLS-DA is an appropriative method to urine groups (18). As a result, the OPLS-DA score plot demonstrated a separation of the groups (Figures 1a–d). These results indicated differences in spectra patterns between the human and animals, which are derived from the differences in metabolites. Therefore, human and animal urine samples can be distinguished by simple statistical analysis after NMR measurement. Figure 1. View largeDownload slide OPLS-DA score plot from the integrated 1H-NMR spectra for the urine samples. (a) Rat versus Human (R2X = 0.41, R2Y = 0.955, Q2 = 0.944), (b) Cat versus Human (R2X = 0.735, R2Y = 0.935, Q2 = 0.89), (c) Dog versus Human (R2X = 0.498, R2Y = 0.57, Q2 = 0.252), (d) Cow versus Human (R2X = 0.504, R2Y = 0.962, Q2 = 0.947), (e) Horse versus Human (R2X = 0.554, R2Y = 0.869, Q2 = 0.773), (F) Monkey versus Human (R2X = 0.649, R2Y = 0.58, Q2 = 0.445), (g) Pig versus Human (R2X = 0.747, R2Y = 0.864, Q2 = 0.71). Figure 1. View largeDownload slide OPLS-DA score plot from the integrated 1H-NMR spectra for the urine samples. (a) Rat versus Human (R2X = 0.41, R2Y = 0.955, Q2 = 0.944), (b) Cat versus Human (R2X = 0.735, R2Y = 0.935, Q2 = 0.89), (c) Dog versus Human (R2X = 0.498, R2Y = 0.57, Q2 = 0.252), (d) Cow versus Human (R2X = 0.504, R2Y = 0.962, Q2 = 0.947), (e) Horse versus Human (R2X = 0.554, R2Y = 0.869, Q2 = 0.773), (F) Monkey versus Human (R2X = 0.649, R2Y = 0.58, Q2 = 0.445), (g) Pig versus Human (R2X = 0.747, R2Y = 0.864, Q2 = 0.71). Then, targeted metabolite profiling method was applied for the acquired NMR spectra. The representative 1H-NMR spectra and difference in metabolites of Human and animal urine are shown Figures 2–8. In rat (Figure 2), 2-oxoglutatate, succinate, acetate and N,N-dimethylglycine were identified as characteristic metabolites differentiated from human metabolites. 1-methylnicotinamide in cats (Figure 3), acetate, lactate, succinate, 1-methylnicotineamide and kynurenate in dogs (Figure 4), acetate, pyruvate, glycine, creatine and benzoate in cows (Figure 5), acetate and benzoate in horse (Figure 6), benzoate in monkeys (Figure 7), acetate, allantonin, hippurate, N-phenylacetylglycine and N-isovaleroylglycine in pigs (Figure 8) were identified as characteristic metabolites of each animal. From theses statistical analysis results, it was found that the NMR pattern difference was derived from the difference in the characteristic metabolites. Figure 2. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Rat urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 2. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Rat urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 3. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Cat urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 3. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Cat urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 4. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Dog urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 4. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Dog urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 5. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Cow urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 5. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Cow urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 6. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Horse urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 6. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Horse urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 7. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Monkey urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 7. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Monkey urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 8. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Pig urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. Figure 8. View largeDownload slide The representative 1H-NMR spectra and difference metabolites of Human and Pig urine. (a) 0.5–5.0 ppm region, (b) 5.0–10.0 ppm region. Stars indicate unknown metabolites. This result can be explained by the difference in physiological metabolism between humans and animals, ruminants and non-ruminants, but it could not be completely explained that the diet was not affected. Urinary constituents in humans and animals may vary depending on the composition, type and time of ingestion of the food. So, the present study has limitations of diet effects, therefore, further studies with a variety of dietary samples, a larger number of urine samples and time series analysis are necessary to offset the effects of diet. Conclusion Our study showed that human and animal urine can be distinguished by 1H-NMR spectra combined with multivariate analysis. Several characteristic metabolites were found and could be used as biomarkers for distinguishing human and animal urine. So, the application of 1H-NMR with multivariate analysis and targeted metabolite profiling analysis might be a valuable tool in differentiating human urine from animal urine. References 1 Castillo-Peinado , L.S. , Luque de Castro , M.D. ( 2016 ) Present and foreseeable future of metabolomics in forensic analysis . Analytica Chimica Acta , 925 , 1 – 15 . Google Scholar Crossref Search ADS PubMed 2 Lood , Y. , Aardal-Eriksson , E. , Webe , C. , Ahlner , J. , Ekman , B. , Wahlberg , J. ( 2018 ) Relationship between testosterone in serum, saliva and urine during treatment with intramuscular testosterone undecanoate in gender dysphoria and male hypogonadism . American Society of Andrology and European Academy of Andrology , 6 , 86 – 93 . 3 Nieddu , M. , Carta , A. , Burrai , L. , Porcu , M.C. , Culeddu , N. , Pirisi , M.A. , et al. . ( 2014 ) Screening method for five commonly used amphetamines in urine by NMR spectroscopy . Applied Magnetic Resonance , 45 , 135 – 144 . Google Scholar Crossref Search ADS 4 Allen , K.R. ( 2011 ) Screening for drugs of abuse: which matrix, oral fluid or urine? Annals of Clinical Biochemistry , 48 , 531 – 541 . Google Scholar Crossref Search ADS PubMed 5 Jaffee , W.B. , Trucco , E. , Levy , S. , Weiss , R.D. ( 2007 ) Is this urine really negative? A systematic review of tampering methods in urine drug screening and testing . Journal of Substance Abuse Treatment , 33 , 33 – 42 . Google Scholar Crossref Search ADS PubMed 6 Cook , J.D. , Caplan , Y.H. , LoDico , C.P. , Bush , D.M. ( 2000 ) The characterization of human urine for specimen validity determination in workplace drug testing: a review . Journal of Analytical Toxicology , 24 , 579 – 588 . Google Scholar Crossref Search ADS PubMed 7 Substance Abuse and Mental Health Services Administration . ( 2015 ) Mandatory guidelines for federal workplace drug testing programs . Federal Register , 80 , 28054 – 28101 . 8 Villena , V.P. ( 2010 ) Beating the system: a study of a creatinine assay and its efficacy in authenticating human urine specimens . Journal of Analytical Toxicology , 34 , 39 – 44 . Google Scholar Crossref Search ADS PubMed 9 FILEP , J. , Földes-Filep , E. ( 1987 ) Effect of urinary pH and urine flow rate on prostaglandin E2 and kallikrein excretion by the conscious dog . The Journal of Physiology , 383 , 1 – 8 . Google Scholar Crossref Search ADS PubMed 10 Cottam , Y.H. , Caley , P. , Wamberg , S. , Hendriks , W.H. ( 2002 ) Feline reference values for urine composition1 . Journal of Nutrition , 132 , 1754 – 1756 . Google Scholar Crossref Search ADS 11 Lehtonen , H.M. , Lindstedt , A. , Järvinen , R. , Sinkkonen , J. , Graça , G. , Viitanen , M. , et al. . ( 2013 ) 1H NMR-based metabolic fingerprinting of urine metabolites after consumption of lingonberries (Vaccinium vitis-idaea) with a high-fat meal . Food Chemistry , 138 , 982 – 990 . Google Scholar Crossref Search ADS PubMed 12 Emwas , A.H. , Salek , R. , Griffin , J. , Merzaban , J. ( 2013 ) NMR-based metabolomics in human disease diagnosis: applications, limitations, and recommendations . Metabolomics: Official Journal of the Metabolomic Society , 9 , 1048 – 1072 . 13 Kim , S. , Yoon , D. , Lee , D. , Yoon , C. , Kim , S. ( 2017 ) 1H-NMR with multivariate analysis for automobile lubricant comparison . Journal of Forensic Science , 62 , 1556 – 4029 . Google Scholar Crossref Search ADS 14 Bharti , S.K. , Roy , R. ( 2012 ) Quantitative 1H NMR spectroscopy . TrAC Trends in Analytical Chemistry , 35 , 5 – 26 . Google Scholar Crossref Search ADS 15 Hong , Y.S. ( 2011 ) NMR-based metabolomics in wine science . Magnetic Resonance in Chemistry , 49 , 13 – 21 . Google Scholar Crossref Search ADS 16 Wheelock , A.M. , Wheelock , C.E. ( 2013 ) Trials and tribulations of ‘omics data analysis: assessing quality of SIMCA-based multivariate models using examples from pulmonary medicine . Molecular Biosystems , 9 , 2589 – 2596 . Google Scholar Crossref Search ADS PubMed 17 Hong , Y.S. , Ahn , Y.T. , Park , J.C. , Lee , J.H. , Lee , H. , Huh , C.S. , et al. . ( 2010 ) 1H NMR-based metabonomic assessment of probiotic effects in a colitis mouse model . Archives of Pharmacal Research , 33 , 1091 – 1101 . Google Scholar Crossref Search ADS PubMed 18 Worley , B. , Powers , R. ( 2013 ) Multivariate analysis in metabolomics . Current Metabolomics , 1 , 92 – 107 . Google Scholar PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Carvalho,, Joana;Rosado,, Tiago;Barroso,, Mário;Gallardo,, Eugenia

doi: 10.1093/jat/bky064pmid: 30192957

Abstract The present work describes the development and validation of an analytical method for the determination of the antiepileptic drugs (AEDs) phenytoin, phenobarbital, carbamazepine and its active metabolite carbamazepine-10,11-epoxide, in oral fluid using liquid chromatography coupled to a diode array detector. Correlation between plasma and oral fluid for these compounds has been proven before, making this matrix a great non-invasive alternative for drug-monitoring purposes. The adaptation of cards, commonly applied in dried blood spots (DBS) sampling, to oral fluid resulted in dried saliva spots (DSS). The extraction procedure consisted of applying 50 μL of oral fluid to WhatmanTM 903 protein saver cards and drying for 1 h. The extraction was performed with 1 mL of acidified methanol (pH = 5.5) for 5 min with agitation. Afterward, the sample was centrifuged for 15 min at 3,500 rpm and the supernatant was evaporated and reconstituted with 80 μL of mobile phase. For the AEDs separation, a Zorbax SB-C18 column (1.8 μm, 4.6 × 250 mm i.d.) maintained at 35°C was used, and the mobile phase consisted of 35% acetonitrile and 65% of water:methanol:triethylamine (75.5%:24.2%:0.3%) isocratically with a flow of 0.5 mL/min, and the wavelength was monitored at 210 nm. The method was validated according to internationally accepted criteria, and linearity between 0.1–10 μg/mL was obtained for all AEDs. This is the first method using DSS for the determination of AEDs, showing great potential for routine use in a laboratory for its simplicity, in addition to the advantages inherent to the use of oral fluid as a sample. Introduction Due to chronic use in treatment of seizure disorders, antiepileptic drugs (AEDs) have received much attention and have been extensively evaluated in therapeutic drug monitoring (TDM) (1). These drugs are known to have a defined serum therapeutic range as well as a measurable and temporally coupled pharmacodynamic end point (i.e. seizures) (1). Although individualization of the dose is essential in epilepsy therapy, identification of the optimal dose on purely clinical grounds can be difficult (2). Seizure control is improved when the active drug species is measured and when serum data are used appropriately (1). Oral fluid has been advocated as an alternative matrix to serum since the 1980s and the increasing interest in free drug concentration monitoring has provided a renewed impetus in oral fluid monitoring of AEDs (2, 3) and its acceptance by clinicians. Several papers have been published reporting the correlations between plasma (serum) and oral fluid. The levels of these AEDs have shown a significant correlation (P < 0.05) (2, 4–9). According to the literature, there is a strong evidence that oral fluid can be used for TDM in patients undergoing epilepsy treatment with carbamazepine (CBZ), phenobarbital (PB) and phenytoin. However, the few data available in the literature for other AEDs such as clonazepam, pregabalin, retigabine, rufinamide, stiripentol, tiagabine, vigabatrin, among others, do not allow their monitoring in oral fluid (10). Routine TDM of carbamazepine-10,11-epoxide (CBZ-EP), a major metabolite of CBZ, is not considered necessary. However, it might be beneficial in patients taking CBZ with other AEDs susceptible to pharmacokinetic interaction, in suspected toxicity and renal insufficiency (11). Studies concluded that measurement of CBZ and CBZ-EP in oral fluid of chronically medicated epileptic patients provides a more reliable estimate of the pharmacodynamically active, free concentrations of these compounds in serum (12). For most patients, the concentration of CBZ-EP at steady state is ~15–20% of the total CBZ concentration. The proportion of epoxide metabolite may be significantly higher in CBZ overdose, and renal failure, or when altered CBZ pharmacokinetics (PK) occur (13). Dried matrix spots (DMSs) sampling is the technique of collecting a dried biological fluid by spotting the liquid specimen onto a collection card and allowing it to dry (14, 15). Recently, DMS technology, especially dried blood spots (DBS), has attracted considerable attention for the bioanalysis of drugs and metabolites from whole blood proving to be a useful tool in drug development (14, 15). Considering the advantages of DBS sampling, including cost savings for sample storage and the possibility of shipping under ambient conditions, dried saliva spots (DSSs) as an alternative to liquid oral fluid for pharmacokinetic evaluation are reported in drug discovery and development literature (14, 15). Abdel-Rehim et al. (16) reported for the first time DSS as sampling technique to determine lidocaine (a local anesthetic drug) by liquid chromatography–mass spectrometry (LC–MS). The authors claim a new dimension in the oral fluid handling process in terms of sampling, storing and transport (16). Adding the easy and non-invasive collection as well as the possibility to be performed with supervision, decreasing the possibility of sample adulteration or substitution (17, 18), this technique can become quite useful regarding AEDs monitoring. The aim of the present work was the development and validation of a simple methodology for the simultaneous determination of selected AEDs in oral fluid samples using DSS in combination with high-performance liquid chromatography system (HPLC)–DAD technology. Materials and Methods Reagents and standards The analytical standards of phenytoin (PHT), PB, CBZ and CBZ-EP were purchased from Sigma-Aldrich (Sintra, Portugal). The internal standard (IS), ketoprofen (KTP) was also purchased from Sigma-Aldrich (Sintra, Portugal). Methanol (Merck Co, Darmstadt, Germany), dichloromethane (Fischer chemical, Loughborough, UK), isopropanol (Fischer chemical, Loughborough, UK), acetonitrile (ACN; Prolabo, Lisbon, Portugal), hexane (Merck Co, Darmstadt, Germany) and ethyl acetate (Fischer chemical, Loughborough, UK) were all of analytical grade. Deionized water was obtained from a Milli-Q System (Millipore, Billerica, MA, USA). Triethylamine (TEA) (Merck Co, Darmstadt, Germany) and formic acid (Panreac Química SA, Barcelona, Spain) were pro-analysis grade. WhatmanTM 903 protein saver cards were acquired from Sigma-Aldrich (Sintra, Portugal). Working solutions were prepared by serial dilutions of stock solutions with methanol to the final concentrations of 200, 20 and 2 μg/mL for PHT, PB, CBZ and CBZ-EP, and a working solution of IS at 5 μg/mL was also prepared in methanol. All solutions were stored in the absence of light at 4°C. Biological specimens Drug-free oral fluid samples used during procedure development, optimization and validation were provided by laboratory staff (CICS, Covilhã, Portugal). Authentic samples were obtained from volunteers undergoing therapy with the studied AEDs. All analyses were carried out according to the ethical standards of the institution and national research committee (the project was approved by the Ethics Committee of the University of Beira Interior, Case No. CE-UBI-Pj-2017-013); the Helsinki declaration and its later amendments or comparable ethical standards were also applied. These samples were collected by spitting and stored refrigerated at −21°C until analysis. HPLC–DAD chromatographic conditions A HPLC 1290 infinity with quaternary pump coupled to a 1290 infinity diode array detector (G4212A DAD) from Agilent technologies (Soquimica, Lisboa, Portugal) was set to perform the chromatographic analysis. All AEDs and IS were separated with a Zorbax SB-C18 (1.8 μm, 4.6 × 250 mm i.d.) analytical column from Agilent Technologies (Soquímica, Lisboa, Portugal). The HPLC–DAD worked on isocratic mode with a mobile phase composed by two solutions. Solution 1 corresponded to 35% of the total mobile phase with ACN, while solution 2 corresponded to the remaining 65% with water:methanol:TEA (75.5:24.2:0.3), pH 6.5. The mobile phase flow rate was 0.5 mL/min and the sampler and column temperatures were set to 4 and 35°C, respectively, with a runtime of 10 min. A sample volume of 50 μL was injected and the analytes were detected at 210 nm, a wavelength described elsewhere for this type of analysis (19–26). Sample preparation The frozen oral fluid samples were allowed to thaw at room temperature and were subsequently centrifuged at 3,500 rpm for 15 min. A 50-μL aliquot of the oral fluid was then placed on a card, where it dried for 1 h. The DSS card was then cut out around the fixed marked diameter and transferred to a clean 15-mL polypropylene falcon centrifuge tube. The liquid extraction of the spot was carried out using 1 mL of methanol acidified with formic acid (pH 5.5). Also 20 μL of IS working solution was added to the extraction solvent. The contents of the falcon tube were agitated at 70 rpm for 5 min on a roller mixer, and the organic phase was transferred to glass tubes and subsequently centrifuged for 15 min at 3,500 rpm. In order to concentrate the analytes, the supernatant was evaporated to dryness under a gentle stream of nitrogen and the dry extracts were reconstituted with 80 μL of mobile phase before being transferred to the autosampler for injection into the HPLC–DAD system. The optimization of this DSS procedure is described in the Results and Discussion section. Validation Procedure The herein described method was fully validated according to the guiding principles of the Food and Drug Administration (FDA) (27), the Scientific Working Group for Forensic Toxicology (SWGTOX) (28) and the specific recommendations of the European Bioanalysis Forum (EBF) for the validation of methodologies involving DBS techniques (29). The evaluated parameters were selectivity, linearity, limits of quantification (LLOQs), limits of detection (LODs), accuracy, precision, recovery, stability and dilution integrity. The selectivity parameter evaluated the presence of potential endogenous interferences (e.g. mineral salts, mucins and digestion enzymes) in oral fluid. This was evaluated analyzing 10 different pools (each pool being a mixture of 4 different oral fluid samples) of AED-free oral fluid samples. If no signals at the retention times of the selected AEDs were detected (<LOD), the method was considered selective. Linearity was tested in the concentration range from LLOQ to 10 μg/mL for all studied AEDs using six calibration standards (0.1, 0.5, 2, 5, 8 and 10 μg/mL). Replicates (n = 5) for each standard were analyzed using the extraction procedure described above during 5 days. Calibration curves were obtained by plotting the peak area ratio between each AED and the IS against concentration. The acceptance criteria included the determination coefficient value (R2) > 0.99 as well as the calibrators’ accuracy within a ±15% (except at the LLOQ, where ±20% was considered acceptable). The LLOQ was considered the lowest concentration measured with adequate precision and accuracy, i.e. with a coefficient of variation (CV, %) of less than 20% and a relative error (RE, %) within ±20% from the nominal concentration. The LOD was estimated using five blank pools of samples from different sources fortified at decreasing concentrations and considered as the lowest concentration yielding a discrete peak clearly distinguishable from the blank with a signal-to-noise (S/N) ratio of at least 3. Inter-day precision of the method was evaluated at a minimum of six concentration levels in fortified oral fluid samples, within a 5-day period. Intra-day precision was determined at three concentration levels (0.1, 2 and 10 μg/mL). For this evaluation, six replicates of blank oral fluid samples spiked with the studied AEDs were analyzed on the same day. Additionally, intermediate precision was determined at three concentration levels covering the predicted linearity range (0.2, 1, and 6 μg/mL). Each sample was analyzed three times over 5 days. The method’s accuracy was characterized in terms of the mean RE between the concentrations measured using the calibration equation and the spiked concentrations; the accepted limit was 15% for all concentrations, except at the LLOQ, where 20% was accepted. The recoveries were obtained at three concentration levels, corresponding to the QCs, using a pool of 10 oral fluid samples from different donors. For each concentration, the recoveries were calculated for each AED by comparing the relative peak area when the AEDs were spiked into the oral fluid sample before extraction (A) with the relative peak area of blank oral fluid samples spiked after the extraction (B). The IS was added to the two sets of sample after extraction. Accordingly, the extraction recoveries = A/B × 100. Stability was assessed on fortified oral fluid aliquots, at QC concentration levels (n = 3) under specific conditions and time intervals (short- and long-term, and freeze/thaw stability). In order to evaluate short-term stability, blank oral fluid samples were spiked and were left at room temperature for 24 h, after which they were applied on cards and processed as previously described. Long-term stability was evaluated by spiking the AED methanol solutions in blank oral fluid samples and applying them on the cards. The DSS were left on a benchtop at room temperature for specific time intervals (t = 1, 2 and 3 weeks). Regarding freeze and thaw stability, oral fluid samples were spiked and stored at −20°C for 24 h. After this period, the frozen samples were thawed unassisted at room temperature, and then refrozen for 12–24 h under the same conditions. This freeze/thaw cycle was repeated twice more, and the samples were processed after the third cycle. The stability evaluation was made by comparing the samples with those prepared and analyzed freshly in the same day. For each stability study, the AED was considered stable if the CV between the two sets of samples was below 15%. Results and Discussion Optimization of the extraction procedure During method development, some laboratories screen different card types and solvent mixtures. This approach allows for the selection of the best combination to facilitate further method optimization (30, 31). The widely used manual extraction method for analyzing DMS samples involves punching a disk from the sample and extracting the analyte with a solvent (typically methanol) containing an appropriate concentration of the IS (32). In fact, the present method does not punch disks from the DSS sample, but instead uses the whole defined circle on the Whatman TM 903 in order to guarantee the full recovery of the 50 μL of sample applied. However, it is important when spotting onto cellulose-based papers to ensure that the sample is uniformly spotted (31), and that it does not spread outside the defined circle spot that will be removed for extraction. Extraction solvent evaluation DMS offers the ability to extract compounds using organic solvents by direct elution of the compound(s) of interest from the cellulose-based cards. This solvent extraction approach has proved highly advantageous in bioanalytical environments where a large number of samples can be processed in a short period of time (31). Experiments have shown that the type of solvent can selectively affect desorption output (33). The extraction solvent must be strong enough to interrupt the binding of analyte to protein in the matrix or the paper material. The analyte of interest is then extracted with gentle shaking or vortexing (34). It has been previously demonstrated that organic solvents such as methanol or ACN precipitated proteins on the paper, leading to a cleaner extract (35). In the present work, a total of nine different organic solvents or mixtures were tested (n = 3): methanol, ACN, methanol:ACN (50:50; v/v), ethyl acetate, dichloromethane, isopropanol, hexane and acidified methanol (pH 5.5) and ACN (pH 5.5). The selection of this pH is in accordance to the reported by Rani et al. (36) that observed an improvement in the extraction efficiency of these drugs. The results obtained when 2 mL of the different extraction solvents were applied to the DSS, and roller mixed during 15 min at 70 rpm, are presented in Figure 1A. An overnight drying time after spotting was applied for this evaluation. Figure 1. View largeDownload slide Effect of extraction solvent on the recovery of AEDs (10 μg/mL) in DSS (A); effect of extraction solvent volume on the recovery of AEDs (10 μg/mL) in DSS (B); effect of extraction time on the recovery of AEDs (10 μg/mL) in DSS (C) and effect of drying time on the recovery of AEDs (10 μg/mL) in DSS (D). Figure 1. View largeDownload slide Effect of extraction solvent on the recovery of AEDs (10 μg/mL) in DSS (A); effect of extraction solvent volume on the recovery of AEDs (10 μg/mL) in DSS (B); effect of extraction time on the recovery of AEDs (10 μg/mL) in DSS (C) and effect of drying time on the recovery of AEDs (10 μg/mL) in DSS (D). It is possible to observe that both methanol and acidified methanol are the solvents that reveal a greater desorption output for all the studied AEDs, allowing good extraction efficiencies (Table III). This is in accordance with the literature that suggests methanol on DBS as the most commonly applied extraction solvent (32). In general, the direct extraction from the cards with hydrophilic organic solvents, such as methanol, avoids the subsequent liquid–liquid extraction (LLE) procedure, which implies a fast sample pretreatment (37). The recoveries obtained for all AEDs decrease significantly when methanol is not applied. Indeed, the fact that methanol is mixed with ACN (50:50; v/v) improves the extraction, resulting in absolute peak areas >5 times greater than those obtained when ACN was used alone. The acidification effect on the extraction solvent is apparently more visible on ACN than methanol. In fact, the absolute peak areas of the AEDs increase >2.5 times when ACN pH 5.5 is applied comparing to pure ACN. Regarding methanol, there is no significant differences when methanol pH 5.5 or pure methanol is used. ANOVA and T-student revealed no significant differences between the two solvents for PB (F(1, 4) = 0.33, P < 0.05), PHT (F(1, 4) = 0.35, P < 0.05), CBZ (F(1, 4) = 0.07, P < 0.05), and CBZ-EP (F(1, 4) = 0.08, P < 0.05). Nevertheless, the extraction with acidified methanol resulted in coefficients of variation typically lower than 11% comparing to those obtained using pure methanol (27–34%). Taking into account the described analysis, acidified methanol was selected as the most appropriate extraction solvent for the DSS procedure. Volume of extraction solvent evaluation As the development of modern sample preparation techniques is focused on miniaturization and facilitation of the sample preparation step, the present procedure uses a reduced volume of sample. However, to be in agreement with the green principles in chemistry, it is also important to reduce the volume of solvents required, always assuring the required sensitivity of the method. A total of four different volumes of acidified methanol (n = 3) were analyzed (1, 2, 3 and 4 mL). The results obtained when the different solvent volumes were applied to the DSS, maintaining the soaking time (roller mixer) of 15 min and agitation of 70 rpm, are presented in Figure 1B. The overnight drying time after spotting was kept for this evaluation. The volume of acidified methanol soaking the DSS appears to have no significant influence on the desorption output for all the studied AEDs. The volume of 1 mL was therefore selected as efficient and quite attractive, especially when compared to the classic sample clean-up procedures, such as solid-phase extraction (SPE) or LLE, which routinely require greater amounts of solvents and steps. Extraction time evaluation Furthermore, it is also important to consider the DSS extraction time by means of the extraction reagent contact time with the DSS in the polypropylene falcon tube. The influence of the soaking time was evaluated by injecting DSS samples after 5, 15, 30 and 60 min (n = 3). The evaluation of more than 1 h of contact seemed unpractical since the main goal of the present work was the development of a simple and fast procedure that could be used routinely. Roller mixer homogenization was kept at 70 rpm. The results obtained (Figure 1C) reveal no significant influence of the extraction time to recover the AEDs present on the DSS. Accordingly, the period of agitation on roller mixer of 5 minute at 70 rpm was considered suitable. Effect of drying time It is considered important to dry DSS samples completely before storage or transport because moisture may harm the specimen by inducing bacterial growth or altering its elution (38, 39). The drying time will depend on the paper type and the applied volume (38). The DSS drying times investigated were 1 and 3 h, and overnight (n = 3). The drying was performed unassisted at room temperature. Although, regarding DBS, immediate drying of the collected blood spot can improve analyte stability and reduce environmental influences (40), in the present work no significant influence was observed for this parameter. The different matrix and studied analytes might also justify the obtained results (Figure 1D). Method Validation Selectivity The selectivity was evaluated by analyzing 10 different oral fluid samples collected from laboratory staff not undergoing AED therapy, and it was checked for interferences of endogenous compounds at the retention times. Other compounds like antidepressants, antipsychotics, different anticonvulsants, caffeine and nicotine metabolites were also checked in order to guarantee the selectivity of the method. No interferences from endogenous substances were observed. Figure 2 shows two chromatograms. The upper corresponds to a DSS sample at the LLOQ processed and analyzed according to the developed method, while the lower is an example of a DSS blank sample for AEDs. Figure 2. View largeDownload slide Selectivity evaluation—comparison of processed DSS at LLOQ (0.1 μg/mL) with a processed DSS blank sample. * Retention times (RT): IS (3.9 min); PB (4.6 min); PHT (6.6 min); CBZ (7.3 min) and CBZ-EP (7.9 min). Figure 2. View largeDownload slide Selectivity evaluation—comparison of processed DSS at LLOQ (0.1 μg/mL) with a processed DSS blank sample. * Retention times (RT): IS (3.9 min); PB (4.6 min); PHT (6.6 min); CBZ (7.3 min) and CBZ-EP (7.9 min). Linearity, limits of detection and quantification The present method was considered linear between 0.1 and 10 μg/mL for all AEDs. To compensate for heterocedasticity, weighted least squares regressions had to be adopted. Among the evaluated weighting factors, 1/x was the one revealing better results for each AED. This evaluation took into account the data obtained during the assessment of the inter-day precision and accuracy and the mean REs. 1/x was the factor that resulted in the lowest sum of errors and simultaneously presented a mean R2 value of at least 0.99. Linear relationships were obtained by means of 1/x weighted least square regression, with calibrators’ accuracy (mean RE (bias) between the measured and spiked concentrations) within a ± 5% interval for all calibrators. Table I shows the calibration data. Table I. Linearity data (n = 5), limits of detection and quantification (n = 10) AEDs Weight Linear range (μg/mL) Linearity R2 LOD (μg/mL) LLOQ (μg/mL) Slopea Intercepta PB 1/x 0.1–10 0.577 ± 0.161 0.176 ± 0.429 0.998 ± 0.001 0.05 0.1 PHT 1/x 0.1–10 0.622 ± 0.380 0.122 ± 0.258 0.998 ± 0.001 0.05 0.1 CBZ 1/x 0.1–10 1.599 ± 0.574 0.070 ± 0.116 0.998 ± 0.002 0.05 0.1 CBZ-EP 1/x 0.1–10 0.543 ± 0.170 0.057 ± 0.097 0.998 ± 0.001 0.05 0.1 AEDs Weight Linear range (μg/mL) Linearity R2 LOD (μg/mL) LLOQ (μg/mL) Slopea Intercepta PB 1/x 0.1–10 0.577 ± 0.161 0.176 ± 0.429 0.998 ± 0.001 0.05 0.1 PHT 1/x 0.1–10 0.622 ± 0.380 0.122 ± 0.258 0.998 ± 0.001 0.05 0.1 CBZ 1/x 0.1–10 1.599 ± 0.574 0.070 ± 0.116 0.998 ± 0.002 0.05 0.1 CBZ-EP 1/x 0.1–10 0.543 ± 0.170 0.057 ± 0.097 0.998 ± 0.001 0.05 0.1 aMean values ± standard deviation. Table I. Linearity data (n = 5), limits of detection and quantification (n = 10) AEDs Weight Linear range (μg/mL) Linearity R2 LOD (μg/mL) LLOQ (μg/mL) Slopea Intercepta PB 1/x 0.1–10 0.577 ± 0.161 0.176 ± 0.429 0.998 ± 0.001 0.05 0.1 PHT 1/x 0.1–10 0.622 ± 0.380 0.122 ± 0.258 0.998 ± 0.001 0.05 0.1 CBZ 1/x 0.1–10 1.599 ± 0.574 0.070 ± 0.116 0.998 ± 0.002 0.05 0.1 CBZ-EP 1/x 0.1–10 0.543 ± 0.170 0.057 ± 0.097 0.998 ± 0.001 0.05 0.1 AEDs Weight Linear range (μg/mL) Linearity R2 LOD (μg/mL) LLOQ (μg/mL) Slopea Intercepta PB 1/x 0.1–10 0.577 ± 0.161 0.176 ± 0.429 0.998 ± 0.001 0.05 0.1 PHT 1/x 0.1–10 0.622 ± 0.380 0.122 ± 0.258 0.998 ± 0.001 0.05 0.1 CBZ 1/x 0.1–10 1.599 ± 0.574 0.070 ± 0.116 0.998 ± 0.002 0.05 0.1 CBZ-EP 1/x 0.1–10 0.543 ± 0.170 0.057 ± 0.097 0.998 ± 0.001 0.05 0.1 aMean values ± standard deviation. The lower limit of quantification (LLOQ) obtained with the present procedure and method was 0.1 μg/mL. The acceptance criteria for LLOQ was set at 20% R.S.D. for precision and ± 20% for bias (41–43). The lowest concentration of each AED in the DSS procedure, that could reliably be differentiated from the background noise, was considered the LOD. The concentration of 0.05 μg/mL was achieved as an LOD. An S/N equal to or greater than 3 was observed (n = 10). These limits can be considered more than satisfactory, taking into account the reduced volume of oral fluid (50 μL) applied on the DSS procedure. The literature applying oral fluid for AEDs determination is not vast, and to best of our knowledge the present work results in the lowest amount of this matrix used for the target AEDs. Previous literature reports the use of salivary sample volumes from 2 to 10 times greater, resulting in LLOQs higher than 0.1 μg/mL (44–48). Although it might not be considered as a significant observation, since this concentration is lower than the therapeutic levels adopted for this matrix, it becomes of great importance to prove the high sensitivity of the simple and fast DSS procedure presented. The herein reported results are only surpassed in the papers by Langel et al. (49), who used 1 mL of oral fluid and LLE coupled to GC–MS, and Ruiz et al. (50), who also used 1 mL of oral fluid and LLE coupled to HPLC–DAD. Those authors reached an LLOQ of 0.02 μg/mL, however the developed multi-methods required a volume of oral fluid 20 times higher than that applied in DSS. Intra-day, inter-day and intermediate precision and accuracy Intra-day precision and accuracy were evaluated by analyzing, on the same day, six replicates of blank oral fluid spiked with the target AEDs at three concentration levels and processed by the described DSS procedure. The CVs obtained were lower than 12% at all studied concentrations, with a mean RE within ±15%. Regarding inter-day precision and accuracy, the study was performed within the 5-day validation period at six concentration levels. The obtained CVs were lower than 13% for all AEDs at the tested concentrations, with an inaccuracy within ±9%. Additionally, intermediate (combined intra- and inter-day) precision was also evaluated using the QC samples preferably near the extremes of the calibration range, but also near the middle (41). The QC samples were prepared (n = 3) and analyzed simultaneously with the calibration curves on 5 different days and resulted in CVs lower than 13% and accuracy within ±5% interval. All data are presented in Table II. Table II. Intra-day, inter-day and intermediate precision and accuracy Inter-day (n = 5) Intra-day (n = 6) Intermediate (n = 15) AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.1 0.099 ± 0.003 3.11 −0.51 0.099 ± 0.009 8.58 −1.20 0.2 0.199 ± 0.015 7.48 −0.63 0.5 0.458 ± 0.022 4.48 −8.50 1 0.992 ± 0,090 8.99 −0.82 2 2.034 ± 0.204 10.22 1.70 2.068 ± 0.200 9.98 3.41 5 5.041 ± 0.304 6.07 0.83 6 6.005 ± 0.658 10.97 0.09 8 7.849 ± 0.363 4.53 −1,89 10 10.085 ± 0.236 2.36 0.85 9.519 ± 0.645 6.45 4.81 PHT 0.1 0.105 ± 0.010 10.28 5.28 0.112 ± 0,007 7.01 12.43 0.2 0.190 ± 0.016 8.17 −4.84 0.5 0.514 ± 0.044 8.85 2.76 1 0.982 ± 0.095 9.48 −1.80 2 1.928 ± 0.211 10.54 −3.59 2.114 ± 0.230 11.52 5.71 5 4.911 ± 0.166 3.33 −1.78 6 5.895 ± 0.633 10.54 −1.76 8 8.023 ± 0.156 1.95 0.28 10 10.167 ± 0.215 2.15 1.67 9.748 ± 0.508 5.08 2.52 CBZ 0.1 0.099 ± 0.009 8.72 −0.51 0.086 ± 0.007 6.91 −14.26 0.2 0.201 ± 0.021 10.32 0.33 0.5 0.543 ± 0.019 3.79 8.58 1 1.005 ± 0.082 8.24 0.55 2 2.053 ± 0.172 8.60 2.66 2.164 ± 0.204 10.2 8.19 5 4.879 ± 0.226 4.53 −2.43 6 6.017 ± 0.660 10.99 0.29 8 8.124 ± 0.350 4.37 1.56 10 9.982 ± 0.505 5.05 −0.18 9.355 ± 0.663 6.63 −6.45 CBZ-EP 0.1 0.099 ± 0.007 6.45 −2.76 0.094 ± 0,004 4.39 −5.96 0.2 0.203 ± 0.024 12.19 1.67 0.5 0.531 ± 0.040 8.40 4.66 1 0.983 ± 0.102 10.19 −1.72 2 2.057 ± 0.213 12.28 2.44 2.109 ± 0.188 9.42 5.43 5 4.949 ± 0.119 2.30 −0.44 6 5.989 ± 0.588 9.79 −0.19 8 8.006 ± 0.281 3.51 0.07 10 9.963 ± 0.350 4.03 −0.52 9.317 ± 0,659 6.59 −6.83 Inter-day (n = 5) Intra-day (n = 6) Intermediate (n = 15) AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.1 0.099 ± 0.003 3.11 −0.51 0.099 ± 0.009 8.58 −1.20 0.2 0.199 ± 0.015 7.48 −0.63 0.5 0.458 ± 0.022 4.48 −8.50 1 0.992 ± 0,090 8.99 −0.82 2 2.034 ± 0.204 10.22 1.70 2.068 ± 0.200 9.98 3.41 5 5.041 ± 0.304 6.07 0.83 6 6.005 ± 0.658 10.97 0.09 8 7.849 ± 0.363 4.53 −1,89 10 10.085 ± 0.236 2.36 0.85 9.519 ± 0.645 6.45 4.81 PHT 0.1 0.105 ± 0.010 10.28 5.28 0.112 ± 0,007 7.01 12.43 0.2 0.190 ± 0.016 8.17 −4.84 0.5 0.514 ± 0.044 8.85 2.76 1 0.982 ± 0.095 9.48 −1.80 2 1.928 ± 0.211 10.54 −3.59 2.114 ± 0.230 11.52 5.71 5 4.911 ± 0.166 3.33 −1.78 6 5.895 ± 0.633 10.54 −1.76 8 8.023 ± 0.156 1.95 0.28 10 10.167 ± 0.215 2.15 1.67 9.748 ± 0.508 5.08 2.52 CBZ 0.1 0.099 ± 0.009 8.72 −0.51 0.086 ± 0.007 6.91 −14.26 0.2 0.201 ± 0.021 10.32 0.33 0.5 0.543 ± 0.019 3.79 8.58 1 1.005 ± 0.082 8.24 0.55 2 2.053 ± 0.172 8.60 2.66 2.164 ± 0.204 10.2 8.19 5 4.879 ± 0.226 4.53 −2.43 6 6.017 ± 0.660 10.99 0.29 8 8.124 ± 0.350 4.37 1.56 10 9.982 ± 0.505 5.05 −0.18 9.355 ± 0.663 6.63 −6.45 CBZ-EP 0.1 0.099 ± 0.007 6.45 −2.76 0.094 ± 0,004 4.39 −5.96 0.2 0.203 ± 0.024 12.19 1.67 0.5 0.531 ± 0.040 8.40 4.66 1 0.983 ± 0.102 10.19 −1.72 2 2.057 ± 0.213 12.28 2.44 2.109 ± 0.188 9.42 5.43 5 4.949 ± 0.119 2.30 −0.44 6 5.989 ± 0.588 9.79 −0.19 8 8.006 ± 0.281 3.51 0.07 10 9.963 ± 0.350 4.03 −0.52 9.317 ± 0,659 6.59 −6.83 All concentrations in μg/mL; CV—coefficient of variation; RE—relative error [(measured concentration−spiked concentration/spiked concentration)] × 100; mean values ± standard deviation. Table II. Intra-day, inter-day and intermediate precision and accuracy Inter-day (n = 5) Intra-day (n = 6) Intermediate (n = 15) AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.1 0.099 ± 0.003 3.11 −0.51 0.099 ± 0.009 8.58 −1.20 0.2 0.199 ± 0.015 7.48 −0.63 0.5 0.458 ± 0.022 4.48 −8.50 1 0.992 ± 0,090 8.99 −0.82 2 2.034 ± 0.204 10.22 1.70 2.068 ± 0.200 9.98 3.41 5 5.041 ± 0.304 6.07 0.83 6 6.005 ± 0.658 10.97 0.09 8 7.849 ± 0.363 4.53 −1,89 10 10.085 ± 0.236 2.36 0.85 9.519 ± 0.645 6.45 4.81 PHT 0.1 0.105 ± 0.010 10.28 5.28 0.112 ± 0,007 7.01 12.43 0.2 0.190 ± 0.016 8.17 −4.84 0.5 0.514 ± 0.044 8.85 2.76 1 0.982 ± 0.095 9.48 −1.80 2 1.928 ± 0.211 10.54 −3.59 2.114 ± 0.230 11.52 5.71 5 4.911 ± 0.166 3.33 −1.78 6 5.895 ± 0.633 10.54 −1.76 8 8.023 ± 0.156 1.95 0.28 10 10.167 ± 0.215 2.15 1.67 9.748 ± 0.508 5.08 2.52 CBZ 0.1 0.099 ± 0.009 8.72 −0.51 0.086 ± 0.007 6.91 −14.26 0.2 0.201 ± 0.021 10.32 0.33 0.5 0.543 ± 0.019 3.79 8.58 1 1.005 ± 0.082 8.24 0.55 2 2.053 ± 0.172 8.60 2.66 2.164 ± 0.204 10.2 8.19 5 4.879 ± 0.226 4.53 −2.43 6 6.017 ± 0.660 10.99 0.29 8 8.124 ± 0.350 4.37 1.56 10 9.982 ± 0.505 5.05 −0.18 9.355 ± 0.663 6.63 −6.45 CBZ-EP 0.1 0.099 ± 0.007 6.45 −2.76 0.094 ± 0,004 4.39 −5.96 0.2 0.203 ± 0.024 12.19 1.67 0.5 0.531 ± 0.040 8.40 4.66 1 0.983 ± 0.102 10.19 −1.72 2 2.057 ± 0.213 12.28 2.44 2.109 ± 0.188 9.42 5.43 5 4.949 ± 0.119 2.30 −0.44 6 5.989 ± 0.588 9.79 −0.19 8 8.006 ± 0.281 3.51 0.07 10 9.963 ± 0.350 4.03 −0.52 9.317 ± 0,659 6.59 −6.83 Inter-day (n = 5) Intra-day (n = 6) Intermediate (n = 15) AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.1 0.099 ± 0.003 3.11 −0.51 0.099 ± 0.009 8.58 −1.20 0.2 0.199 ± 0.015 7.48 −0.63 0.5 0.458 ± 0.022 4.48 −8.50 1 0.992 ± 0,090 8.99 −0.82 2 2.034 ± 0.204 10.22 1.70 2.068 ± 0.200 9.98 3.41 5 5.041 ± 0.304 6.07 0.83 6 6.005 ± 0.658 10.97 0.09 8 7.849 ± 0.363 4.53 −1,89 10 10.085 ± 0.236 2.36 0.85 9.519 ± 0.645 6.45 4.81 PHT 0.1 0.105 ± 0.010 10.28 5.28 0.112 ± 0,007 7.01 12.43 0.2 0.190 ± 0.016 8.17 −4.84 0.5 0.514 ± 0.044 8.85 2.76 1 0.982 ± 0.095 9.48 −1.80 2 1.928 ± 0.211 10.54 −3.59 2.114 ± 0.230 11.52 5.71 5 4.911 ± 0.166 3.33 −1.78 6 5.895 ± 0.633 10.54 −1.76 8 8.023 ± 0.156 1.95 0.28 10 10.167 ± 0.215 2.15 1.67 9.748 ± 0.508 5.08 2.52 CBZ 0.1 0.099 ± 0.009 8.72 −0.51 0.086 ± 0.007 6.91 −14.26 0.2 0.201 ± 0.021 10.32 0.33 0.5 0.543 ± 0.019 3.79 8.58 1 1.005 ± 0.082 8.24 0.55 2 2.053 ± 0.172 8.60 2.66 2.164 ± 0.204 10.2 8.19 5 4.879 ± 0.226 4.53 −2.43 6 6.017 ± 0.660 10.99 0.29 8 8.124 ± 0.350 4.37 1.56 10 9.982 ± 0.505 5.05 −0.18 9.355 ± 0.663 6.63 −6.45 CBZ-EP 0.1 0.099 ± 0.007 6.45 −2.76 0.094 ± 0,004 4.39 −5.96 0.2 0.203 ± 0.024 12.19 1.67 0.5 0.531 ± 0.040 8.40 4.66 1 0.983 ± 0.102 10.19 −1.72 2 2.057 ± 0.213 12.28 2.44 2.109 ± 0.188 9.42 5.43 5 4.949 ± 0.119 2.30 −0.44 6 5.989 ± 0.588 9.79 −0.19 8 8.006 ± 0.281 3.51 0.07 10 9.963 ± 0.350 4.03 −0.52 9.317 ± 0,659 6.59 −6.83 All concentrations in μg/mL; CV—coefficient of variation; RE—relative error [(measured concentration−spiked concentration/spiked concentration)] × 100; mean values ± standard deviation. Extraction efficiency Lower recovery values are usually acceptable as long as data for LLOQ, LOD, precision and accuracy (bias) are adequate (41, 42, 51, 52). For this determination, two sets of samples (n = 3) were prepared by spiking blank oral fluid with the target AEDs at the QC concentration levels. Set 1 represented post-DSS procedure spikes (representing 100% recovery), while set 2 consisted of pre-DSS procedure spikes. The recovery results were obtained by comparison of peak area ratios of sample set 2 with those of the corresponding peaks in sample set 1, and ranged from 41% to 61% (Table III), which we found acceptable for this method. Table III. Recovery (%) of the studied AEDs under the optimized DSS procedure (n = 3) AEDs Spiked (μg/mL) Recovery (%)ª PB 0.2 50.6 ± 6.5 1 47.7 ± 4.7 6 50.6 ± 4.4 PHT 0.2 60.9 ± 3.5 1 58.5 ± 2.7 6 52.6 ± 4.5 CBZ 0.2 45.5 ± 0.0 1 40.8 ± 2.1 6 43.9 ± 5.2 CBZ-EP 0.2 49.4 ± 2.5 1 43.7 ± 4.9 6 49.1 ± 5.9 AEDs Spiked (μg/mL) Recovery (%)ª PB 0.2 50.6 ± 6.5 1 47.7 ± 4.7 6 50.6 ± 4.4 PHT 0.2 60.9 ± 3.5 1 58.5 ± 2.7 6 52.6 ± 4.5 CBZ 0.2 45.5 ± 0.0 1 40.8 ± 2.1 6 43.9 ± 5.2 CBZ-EP 0.2 49.4 ± 2.5 1 43.7 ± 4.9 6 49.1 ± 5.9 ªMean values ± standard deviation. Table III. Recovery (%) of the studied AEDs under the optimized DSS procedure (n = 3) AEDs Spiked (μg/mL) Recovery (%)ª PB 0.2 50.6 ± 6.5 1 47.7 ± 4.7 6 50.6 ± 4.4 PHT 0.2 60.9 ± 3.5 1 58.5 ± 2.7 6 52.6 ± 4.5 CBZ 0.2 45.5 ± 0.0 1 40.8 ± 2.1 6 43.9 ± 5.2 CBZ-EP 0.2 49.4 ± 2.5 1 43.7 ± 4.9 6 49.1 ± 5.9 AEDs Spiked (μg/mL) Recovery (%)ª PB 0.2 50.6 ± 6.5 1 47.7 ± 4.7 6 50.6 ± 4.4 PHT 0.2 60.9 ± 3.5 1 58.5 ± 2.7 6 52.6 ± 4.5 CBZ 0.2 45.5 ± 0.0 1 40.8 ± 2.1 6 43.9 ± 5.2 CBZ-EP 0.2 49.4 ± 2.5 1 43.7 ± 4.9 6 49.1 ± 5.9 ªMean values ± standard deviation. Indeed, the recovery values appear to be lower than the reported literature. Most authors using this biological specimen for AEDs determination apply the classic preconcentration and sample clean-up techniques, such as SPE (47), protein precipitation (44) and LLE (45, 48–50, 53). These techniques are well settled in routine laboratory work and have proven to result in extraction efficiencies above 90%. Miniaturized procedures for the same purpose have not been reported yet, and few studies with DSS are available. However, Abdel-Rehim et al. (16) reported that the DSS analysis of the spiked oral fluid sample compared with direct injection of the same concentration of lidocaine resulted in a recovery near to 100%. The different cards used and drug analysis among many other factors may explain different extraction recoveries. Stability The stability of the analytes during the whole analytical procedure is a prerequisite for reliable quantification (41). All data are presented in Table IV. Short-term stability was assessed with blank oral fluid samples spiked at QC concentration levels and left at room temperature for 24 h previous to sampling on the DSS cards. The samples revealed great stability in benchtop conditions during a 24-h period. The CVs obtained were lower than 15% at all studied concentrations, with a mean RE within ± 11%. Table IV. Short-term, freeze/thaw and long-term stability and accuracy (n = 3) for oral fluid samples Short-term (n = 3) Freeze/thaw (n = 3) Long-term stability (n = 3) 1 week 2 weeks 3 weeks AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.2 0.205 ± 0.015 7.69 2.53 0.212 ± 0.010 4.13 6.43 0.202 ± 0.015 6.14 0.95 0.194 ± 0.021 8.64 −3.16 0.184 ± 0.015 5.97 −8.13 1 0.897 ± 0.050 4.99 −10.33 0.929 ± 0.048 3.97 −7.07 0.949 ± 0.035 2.88 −5.14 1.017 ± 0.116 9.49 1.73 0.990 ± 0.113 11.44 −1.01 6 5.648 ± 0.453 7.56 −5.86 5.930 ± 0.731 9.95 −1.16 5.875 ± 0.672 9.14 −2.08 5.910 ± 0.463 6.31 −1.50 5.943 ± 0.295 4.01 −0.95 PHT 0.2 0.190 ± 0.012 5.85 −5.12 0.213 ± 0.001 0.31 6.26 0.202 ± 0.018 7.33 0.84 0.202 ± 0.025 10.31 1.24 0.205 ± 0.016 6.73 2.27 1 1.065 ± 0.074 7.40 6.51 0.936 ± 0.104 8.51 −6.43 0.965 ± 0.058 4.71 −3.51 0.953 ± 0.051 4.13 −4.70 1.045 ± 0.088 6.23 4.45 6 5.979 ± 0.293 4.88 −0.35 6.265 ± 0.501 6.82 4.41 6.128 ± 0.218 2.97 2.13 6.618 ± 0.198 2.70 10.30 6.415 ± 0.551 7.50 6.92 CBZ 0.2 0.197 ± 0.029 14.58 −1.64 0.205 ± 0.005 1.91 2.32 0.207 ± 0.014 5.52 3.48 0.187 ± 0.013 5.20 −6.60 0.193 ± 0.016 6.40 −3.57 1 0.990 ± 0.097 9.65 −1.03 1.070 ± 0.082 6.71 6.96 0.980 ± 0.101 8.23 −1.98 0.900 ± 0.053 4.32 −10.01 0.900 ± 0.020 1.40 −9.96 6 5.823 ± 0.427 7.11 −2.95 5.711 ± 0.532 7.24 −4.82 5.214 ± 0.120 1.63 −13.10 5.491 ± 0.299 4.07 −8.49 5.199 ± 0.054 0.73 −13.36 CBZ-EP 0.2 0.202 ± 0.025 12.46 0.78 0.204 ± 0.016 6.71 1.85 0.204 ± 0.002 0.78 2.06 0.199 ± 0.011 4.56 −0.61 0.203 ± 0.019 7.70 1.34 1 0.987 ± 0.107 10.66 −1.26 1.094 ± 0.041 3.34 9.44 1.069 ± 0.027 2.24 6.86 1.056 ± 0.048 3.96 5.57 1.061 ± 0.073 5.17 6.13 6 5.389 ± 0.188 3.14 −10.18 6.574 ± 0.228 3.10 9.57 6.524 ± 0.305 4.15 8.74 6.660 ± 0.191 2.61 11.00 5.924 ± 0.608 8.27 −1.26 Short-term (n = 3) Freeze/thaw (n = 3) Long-term stability (n = 3) 1 week 2 weeks 3 weeks AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.2 0.205 ± 0.015 7.69 2.53 0.212 ± 0.010 4.13 6.43 0.202 ± 0.015 6.14 0.95 0.194 ± 0.021 8.64 −3.16 0.184 ± 0.015 5.97 −8.13 1 0.897 ± 0.050 4.99 −10.33 0.929 ± 0.048 3.97 −7.07 0.949 ± 0.035 2.88 −5.14 1.017 ± 0.116 9.49 1.73 0.990 ± 0.113 11.44 −1.01 6 5.648 ± 0.453 7.56 −5.86 5.930 ± 0.731 9.95 −1.16 5.875 ± 0.672 9.14 −2.08 5.910 ± 0.463 6.31 −1.50 5.943 ± 0.295 4.01 −0.95 PHT 0.2 0.190 ± 0.012 5.85 −5.12 0.213 ± 0.001 0.31 6.26 0.202 ± 0.018 7.33 0.84 0.202 ± 0.025 10.31 1.24 0.205 ± 0.016 6.73 2.27 1 1.065 ± 0.074 7.40 6.51 0.936 ± 0.104 8.51 −6.43 0.965 ± 0.058 4.71 −3.51 0.953 ± 0.051 4.13 −4.70 1.045 ± 0.088 6.23 4.45 6 5.979 ± 0.293 4.88 −0.35 6.265 ± 0.501 6.82 4.41 6.128 ± 0.218 2.97 2.13 6.618 ± 0.198 2.70 10.30 6.415 ± 0.551 7.50 6.92 CBZ 0.2 0.197 ± 0.029 14.58 −1.64 0.205 ± 0.005 1.91 2.32 0.207 ± 0.014 5.52 3.48 0.187 ± 0.013 5.20 −6.60 0.193 ± 0.016 6.40 −3.57 1 0.990 ± 0.097 9.65 −1.03 1.070 ± 0.082 6.71 6.96 0.980 ± 0.101 8.23 −1.98 0.900 ± 0.053 4.32 −10.01 0.900 ± 0.020 1.40 −9.96 6 5.823 ± 0.427 7.11 −2.95 5.711 ± 0.532 7.24 −4.82 5.214 ± 0.120 1.63 −13.10 5.491 ± 0.299 4.07 −8.49 5.199 ± 0.054 0.73 −13.36 CBZ-EP 0.2 0.202 ± 0.025 12.46 0.78 0.204 ± 0.016 6.71 1.85 0.204 ± 0.002 0.78 2.06 0.199 ± 0.011 4.56 −0.61 0.203 ± 0.019 7.70 1.34 1 0.987 ± 0.107 10.66 −1.26 1.094 ± 0.041 3.34 9.44 1.069 ± 0.027 2.24 6.86 1.056 ± 0.048 3.96 5.57 1.061 ± 0.073 5.17 6.13 6 5.389 ± 0.188 3.14 −10.18 6.574 ± 0.228 3.10 9.57 6.524 ± 0.305 4.15 8.74 6.660 ± 0.191 2.61 11.00 5.924 ± 0.608 8.27 −1.26 All concentrations in μg/mL; CV—coefficient of variation; RE—relative error [(measured concentration−spiked concentration/spiked concentration)] × 100; Mean values ± standard deviation. Table IV. Short-term, freeze/thaw and long-term stability and accuracy (n = 3) for oral fluid samples Short-term (n = 3) Freeze/thaw (n = 3) Long-term stability (n = 3) 1 week 2 weeks 3 weeks AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.2 0.205 ± 0.015 7.69 2.53 0.212 ± 0.010 4.13 6.43 0.202 ± 0.015 6.14 0.95 0.194 ± 0.021 8.64 −3.16 0.184 ± 0.015 5.97 −8.13 1 0.897 ± 0.050 4.99 −10.33 0.929 ± 0.048 3.97 −7.07 0.949 ± 0.035 2.88 −5.14 1.017 ± 0.116 9.49 1.73 0.990 ± 0.113 11.44 −1.01 6 5.648 ± 0.453 7.56 −5.86 5.930 ± 0.731 9.95 −1.16 5.875 ± 0.672 9.14 −2.08 5.910 ± 0.463 6.31 −1.50 5.943 ± 0.295 4.01 −0.95 PHT 0.2 0.190 ± 0.012 5.85 −5.12 0.213 ± 0.001 0.31 6.26 0.202 ± 0.018 7.33 0.84 0.202 ± 0.025 10.31 1.24 0.205 ± 0.016 6.73 2.27 1 1.065 ± 0.074 7.40 6.51 0.936 ± 0.104 8.51 −6.43 0.965 ± 0.058 4.71 −3.51 0.953 ± 0.051 4.13 −4.70 1.045 ± 0.088 6.23 4.45 6 5.979 ± 0.293 4.88 −0.35 6.265 ± 0.501 6.82 4.41 6.128 ± 0.218 2.97 2.13 6.618 ± 0.198 2.70 10.30 6.415 ± 0.551 7.50 6.92 CBZ 0.2 0.197 ± 0.029 14.58 −1.64 0.205 ± 0.005 1.91 2.32 0.207 ± 0.014 5.52 3.48 0.187 ± 0.013 5.20 −6.60 0.193 ± 0.016 6.40 −3.57 1 0.990 ± 0.097 9.65 −1.03 1.070 ± 0.082 6.71 6.96 0.980 ± 0.101 8.23 −1.98 0.900 ± 0.053 4.32 −10.01 0.900 ± 0.020 1.40 −9.96 6 5.823 ± 0.427 7.11 −2.95 5.711 ± 0.532 7.24 −4.82 5.214 ± 0.120 1.63 −13.10 5.491 ± 0.299 4.07 −8.49 5.199 ± 0.054 0.73 −13.36 CBZ-EP 0.2 0.202 ± 0.025 12.46 0.78 0.204 ± 0.016 6.71 1.85 0.204 ± 0.002 0.78 2.06 0.199 ± 0.011 4.56 −0.61 0.203 ± 0.019 7.70 1.34 1 0.987 ± 0.107 10.66 −1.26 1.094 ± 0.041 3.34 9.44 1.069 ± 0.027 2.24 6.86 1.056 ± 0.048 3.96 5.57 1.061 ± 0.073 5.17 6.13 6 5.389 ± 0.188 3.14 −10.18 6.574 ± 0.228 3.10 9.57 6.524 ± 0.305 4.15 8.74 6.660 ± 0.191 2.61 11.00 5.924 ± 0.608 8.27 −1.26 Short-term (n = 3) Freeze/thaw (n = 3) Long-term stability (n = 3) 1 week 2 weeks 3 weeks AEDs Spiked Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) Measured CV (%) RE (%) PB 0.2 0.205 ± 0.015 7.69 2.53 0.212 ± 0.010 4.13 6.43 0.202 ± 0.015 6.14 0.95 0.194 ± 0.021 8.64 −3.16 0.184 ± 0.015 5.97 −8.13 1 0.897 ± 0.050 4.99 −10.33 0.929 ± 0.048 3.97 −7.07 0.949 ± 0.035 2.88 −5.14 1.017 ± 0.116 9.49 1.73 0.990 ± 0.113 11.44 −1.01 6 5.648 ± 0.453 7.56 −5.86 5.930 ± 0.731 9.95 −1.16 5.875 ± 0.672 9.14 −2.08 5.910 ± 0.463 6.31 −1.50 5.943 ± 0.295 4.01 −0.95 PHT 0.2 0.190 ± 0.012 5.85 −5.12 0.213 ± 0.001 0.31 6.26 0.202 ± 0.018 7.33 0.84 0.202 ± 0.025 10.31 1.24 0.205 ± 0.016 6.73 2.27 1 1.065 ± 0.074 7.40 6.51 0.936 ± 0.104 8.51 −6.43 0.965 ± 0.058 4.71 −3.51 0.953 ± 0.051 4.13 −4.70 1.045 ± 0.088 6.23 4.45 6 5.979 ± 0.293 4.88 −0.35 6.265 ± 0.501 6.82 4.41 6.128 ± 0.218 2.97 2.13 6.618 ± 0.198 2.70 10.30 6.415 ± 0.551 7.50 6.92 CBZ 0.2 0.197 ± 0.029 14.58 −1.64 0.205 ± 0.005 1.91 2.32 0.207 ± 0.014 5.52 3.48 0.187 ± 0.013 5.20 −6.60 0.193 ± 0.016 6.40 −3.57 1 0.990 ± 0.097 9.65 −1.03 1.070 ± 0.082 6.71 6.96 0.980 ± 0.101 8.23 −1.98 0.900 ± 0.053 4.32 −10.01 0.900 ± 0.020 1.40 −9.96 6 5.823 ± 0.427 7.11 −2.95 5.711 ± 0.532 7.24 −4.82 5.214 ± 0.120 1.63 −13.10 5.491 ± 0.299 4.07 −8.49 5.199 ± 0.054 0.73 −13.36 CBZ-EP 0.2 0.202 ± 0.025 12.46 0.78 0.204 ± 0.016 6.71 1.85 0.204 ± 0.002 0.78 2.06 0.199 ± 0.011 4.56 −0.61 0.203 ± 0.019 7.70 1.34 1 0.987 ± 0.107 10.66 −1.26 1.094 ± 0.041 3.34 9.44 1.069 ± 0.027 2.24 6.86 1.056 ± 0.048 3.96 5.57 1.061 ± 0.073 5.17 6.13 6 5.389 ± 0.188 3.14 −10.18 6.574 ± 0.228 3.10 9.57 6.524 ± 0.305 4.15 8.74 6.660 ± 0.191 2.61 11.00 5.924 ± 0.608 8.27 −1.26 All concentrations in μg/mL; CV—coefficient of variation; RE—relative error [(measured concentration−spiked concentration/spiked concentration)] × 100; Mean values ± standard deviation. To evaluate freeze–thaw stability, the QC samples were submitted to three cycles, completely thawed and refrozen for 24 h between cycles at the anticipated temperature of sample storage (54). This was also performed previously to sampling on the DSS cards, also reporting to be stable under these conditions for all AEDs and concentration levels. The obtained CVs were lower than 10% for all AEDs at the tested concentrations, with an inaccuracy within ±14%. It is accepted that the cards matrix stabilizes most analytes in DMSs, but the rate of sample degradation will vary by analyte. Stability should also be evaluated prior to sample collection because this has direct implications for sample handling and storage (55). For long-term stability, the sample matrix should be stored under storage conditions, i.e. in the same vessels, at the same temperature and over a storage period at least as long as the one expected for authentic samples (41, 42, 51, 56). In the present work, after spiking the AEDs methanol solutions in blank oral fluid, the prepared QC concentration levels were applied on the cards and left on benchtop unassisted for specific time intervals (t = 1, 2 and 3 weeks). During these periods of assessment, the CVs obtained were lower than 12% at all studied concentrations, with a mean RE within ±14%. This is of great importance, since one of the main advantages pointed at DMS sampling procedures is its high stability. The present procedure ensures that the sampled DSS can be stored at room temperature for a minimum period of 3 weeks prior to analysis, allowing also the easy shipping of the sample during this period if necessary. No drying with desiccant material or refrigeration conditions are needed during this interval of time which also reduces the costs associated. Dilution integrity As it might be expected that some sample concentrations might exceed the upper limit of quantitation, a test for sample dilution with blank oral fluid during validation was performed. It is in agreement that one or more additional QC samples at concentrations several times higher than the upper limit of the calibration curve should be prepared, covering the maximum expected dilution (54). The dilution integrity was assessed by diluting with blank oral fluid (1:2, 1:5 and 1:10) to bring the concentration to two concentration levels (20 and 60 μg/mL). The measured concentration was compared to the concentration spiked of freshly prepared samples and processed as DSS. Table V shows the dilution integrity for all tested dilutions and assessed concentrations. The obtained CVs were lower than 15% for all AEDs, with an inaccuracy within ±9%. Table V. Dilution integrity (n = 3) for oral fluid samples in DSS Dilution factor 1:2 1:5 1:10 AEDs Concentrations Measured CV (%) RE (%) Measured CV (%) BIAS (%) Measured CV (%) BIAS (%) PB 20 19.791 ± 0.05 0.24 −1.05 19.273 ± 0.057 0.30 −3.64 19.731 ± 0.142 0.72 −1.34 60 57.382 ± 0.54 0.93 −4.36 57.374 ± 0.378 0.66 −4.38 65.023 ± 0.182 0.28 8.37 PHT 20 19.475 ± 0.12 0.63 −2.62 19.551 ± 0.095 0.49 −2.25 19.892 ± 0.087 0.44 −0.54 60 59.982 ± 0.39 0.64 −0.03 61.429 ± 0.831 1.35 2.38 56.323 ± 0.159 0.28 −6.13 CBZ 20 20.683±4E−3 0.02 3.42 20.223 ± 0.153 0.76 1.11 19.280 ± 0.062 0.32 −3.60 60 57.292 ± 0.28 0.49 −4.51 57.580 ± 0.629 1.09 −4.03 55.346 ± 0.142 0.26 −7.76 CBZ-EP 20 18.813 ± 0.02 0.09 −5.94 20.172 ± 0.103 0.51 0.86 20.257 ± 0.007 0.03 1.29 60 57.793 ± 0.40 0.69 −3.68 59.053 ± 0.573 0.97 −1.58 63.341 ± 0.202 0.32 5.57 Dilution factor 1:2 1:5 1:10 AEDs Concentrations Measured CV (%) RE (%) Measured CV (%) BIAS (%) Measured CV (%) BIAS (%) PB 20 19.791 ± 0.05 0.24 −1.05 19.273 ± 0.057 0.30 −3.64 19.731 ± 0.142 0.72 −1.34 60 57.382 ± 0.54 0.93 −4.36 57.374 ± 0.378 0.66 −4.38 65.023 ± 0.182 0.28 8.37 PHT 20 19.475 ± 0.12 0.63 −2.62 19.551 ± 0.095 0.49 −2.25 19.892 ± 0.087 0.44 −0.54 60 59.982 ± 0.39 0.64 −0.03 61.429 ± 0.831 1.35 2.38 56.323 ± 0.159 0.28 −6.13 CBZ 20 20.683±4E−3 0.02 3.42 20.223 ± 0.153 0.76 1.11 19.280 ± 0.062 0.32 −3.60 60 57.292 ± 0.28 0.49 −4.51 57.580 ± 0.629 1.09 −4.03 55.346 ± 0.142 0.26 −7.76 CBZ-EP 20 18.813 ± 0.02 0.09 −5.94 20.172 ± 0.103 0.51 0.86 20.257 ± 0.007 0.03 1.29 60 57.793 ± 0.40 0.69 −3.68 59.053 ± 0.573 0.97 −1.58 63.341 ± 0.202 0.32 5.57 All concentrations in μg/mL; CV—coefficient of variation; RE—relative error [(measured concentration−spiked concentration/spiked concentration)] × 100; mean values ± standard deviation. Table V. Dilution integrity (n = 3) for oral fluid samples in DSS Dilution factor 1:2 1:5 1:10 AEDs Concentrations Measured CV (%) RE (%) Measured CV (%) BIAS (%) Measured CV (%) BIAS (%) PB 20 19.791 ± 0.05 0.24 −1.05 19.273 ± 0.057 0.30 −3.64 19.731 ± 0.142 0.72 −1.34 60 57.382 ± 0.54 0.93 −4.36 57.374 ± 0.378 0.66 −4.38 65.023 ± 0.182 0.28 8.37 PHT 20 19.475 ± 0.12 0.63 −2.62 19.551 ± 0.095 0.49 −2.25 19.892 ± 0.087 0.44 −0.54 60 59.982 ± 0.39 0.64 −0.03 61.429 ± 0.831 1.35 2.38 56.323 ± 0.159 0.28 −6.13 CBZ 20 20.683±4E−3 0.02 3.42 20.223 ± 0.153 0.76 1.11 19.280 ± 0.062 0.32 −3.60 60 57.292 ± 0.28 0.49 −4.51 57.580 ± 0.629 1.09 −4.03 55.346 ± 0.142 0.26 −7.76 CBZ-EP 20 18.813 ± 0.02 0.09 −5.94 20.172 ± 0.103 0.51 0.86 20.257 ± 0.007 0.03 1.29 60 57.793 ± 0.40 0.69 −3.68 59.053 ± 0.573 0.97 −1.58 63.341 ± 0.202 0.32 5.57 Dilution factor 1:2 1:5 1:10 AEDs Concentrations Measured CV (%) RE (%) Measured CV (%) BIAS (%) Measured CV (%) BIAS (%) PB 20 19.791 ± 0.05 0.24 −1.05 19.273 ± 0.057 0.30 −3.64 19.731 ± 0.142 0.72 −1.34 60 57.382 ± 0.54 0.93 −4.36 57.374 ± 0.378 0.66 −4.38 65.023 ± 0.182 0.28 8.37 PHT 20 19.475 ± 0.12 0.63 −2.62 19.551 ± 0.095 0.49 −2.25 19.892 ± 0.087 0.44 −0.54 60 59.982 ± 0.39 0.64 −0.03 61.429 ± 0.831 1.35 2.38 56.323 ± 0.159 0.28 −6.13 CBZ 20 20.683±4E−3 0.02 3.42 20.223 ± 0.153 0.76 1.11 19.280 ± 0.062 0.32 −3.60 60 57.292 ± 0.28 0.49 −4.51 57.580 ± 0.629 1.09 −4.03 55.346 ± 0.142 0.26 −7.76 CBZ-EP 20 18.813 ± 0.02 0.09 −5.94 20.172 ± 0.103 0.51 0.86 20.257 ± 0.007 0.03 1.29 60 57.793 ± 0.40 0.69 −3.68 59.053 ± 0.573 0.97 −1.58 63.341 ± 0.202 0.32 5.57 All concentrations in μg/mL; CV—coefficient of variation; RE—relative error [(measured concentration−spiked concentration/spiked concentration)] × 100; mean values ± standard deviation. Method Applicability The method was successfully applied in routine analysis of the target AEDs in oral fluid by sampling in DSS. The authentic samples were provided by different volunteers undergoing AED therapy with CBZ, PHT or PB. After drying the samples in DSS, they were treated and analyzed according to the procedure previously described. The concentrations ranged from 0.37 to 2.08 μg/mL for CBZ and 0.94 to 1.79 μg/mL for PHT. Regarding PB, only one authentic sample was analyzed (3.49 μg/mL). Figure 3 presents chromatograms of five authentic samples and the concentrations found (no dilutions were made). Figure 3. View largeDownload slide Chromatograms of five authentic samples reporting the concentrations of the AEDs in oral fluid. Figure 3. View largeDownload slide Chromatograms of five authentic samples reporting the concentrations of the AEDs in oral fluid. Conclusions The present analytical method was fully validated, showing to be selective, precise and accurate for the determination of the target AEDs (PB, PHT, CBZ and its active metabolite CBZ-EP) in oral fluid using DSS as a sample preparation technique. The extraction procedure was optimized, achieving an LOD and LLOQ of 0.05 and 0.1 μg/mL, respectively, with mean recoveries ranging from 41% to 61%. This is the first time DSS is used as a sampling technique for determination of AEDs in oral fluid samples revealing great advantages related to specimen collection, simplicity and proving to be a user- and environmental-friendly procedure. In order to implement routinely in laboratory analysis, it should be compared to traditional approaches, particularly taking into account the low volume of sample that was used. The herein described procedure results in a promising alternative for AEDs monitoring, it is non-invasive and allows sample stability for at least 3 weeks on benchtop. To the best of our knowledge, the present work also reports the lowest volume of oral fluid (50 μL) required to determine these AEDs, which can be quite advantageous in case of poor oral fluid production. Acknowledgments We also acknowledge the Farmácias do Grupo Holon-Covilhã for allowing us to kindly collect samples at their facilities, and Farmácia Crespo in Covilhã for kindly providing us other anticonvulsant substances. Funding The authors acknowledge the Fundação para a Ciência e Tecnologia (FCT) (Project UID/Multi /00709/2013) and the POCI-COMPETE 2020—Operational Programme Competitiveness and Internationalisation in Axis I—Strengthening research, technological development and innovation (Project POCI-01-0145-FEDER-007491). References 1 Langman , L.J. ( 2007 ) The use of oral fluid for therapeutic drug management . Annals of the New York Academy of Sciences , 1098 , 145 – 166 . Google Scholar Crossref Search ADS PubMed 2 Patsalos , P.N. , Berry , D.J. , Bourgeois , B.F. , Cloyd , J.C. , Glauser , T.A. , Johannessen , S.I. , et al. . ( 2008 ) Antiepileptic drugs—best practice guidelines for therapeutic drug monitoring: a position paper by the subcommission on therapeutic drug monitoring, ILAE Commission on Therapeutic Strategies . Epilepsia , 49 , 1239 – 1276 . Google Scholar Crossref Search ADS PubMed 3 Liu , H. , Delgado , M.R. ( 1999 ) Therapeutic drug concentration monitoring using saliva samples. Focus on anticonvulsants . Clinical Pharmacokinetics , 36 , 453 – 470 . Google Scholar Crossref Search ADS PubMed 4 Kaewdoung , P. , Puripokai , C. , Tantisira , M.H. , Laanwprasert , S. ( 2016 ) Relationship between carbamazepine concentrations in serum and saliva of Thai epileptic patients . Thai Journal of Pharmaceutical Sciences (TJPS) , 39 , 1 – 34 . 5 Dwivedi , R. , Singh , M. , Kaleekal , T. , Gupta , Y.K. , Tripathi , M. ( 2016 ) Concentration of antiepileptic drugs in persons with epilepsy: a comparative study in serum and saliva . International Journal of Neuroscience , 126 , 972 – 978 . Google Scholar Crossref Search ADS PubMed 6 André , P. , Novy , J. , Decosterd , L.A. , Buclin , T. , Rothuizen , L.E. ( 2015 ) Therapeutic drug monitoring of antiepileptic drugs in the 21st century . Epileptologie , 32 , 78 – 84 . 7 Vasudev , A. , Tripathi , K. , Puri , V. ( 2002 ) Correlation of serum and salivary carbamazepine concentration in epileptic patients: implications for therapeutic drug monitoring . Neurology India , 50 , 60 – 62 . Google Scholar PubMed 8 Herkes , G. , Eadie , M. ( 1990 ) Possible roles for frequent salivary antiepileptic drug monitoring in the management of epilepsy . Epilepsy Research , 6 , 146 – 154 . Google Scholar Crossref Search ADS PubMed 9 Liamsuwan , S. , Jaiweerawattana , U. ( 2011 ) Correlation between serum and salivary phenytoin concentrations in Thai epileptic children . Journal of the Medical Association of Thailand=Chotmaihet thangphaet , 94 , S172 – S177 . Google Scholar PubMed 10 Patsalos , P.N. , Berry , D.J. ( 2013 ) Therapeutic drug monitoring of antiepileptic drugs by use of saliva . Therapeutic Drug Monitoring , 35 , 4 – 29 . Google Scholar Crossref Search ADS PubMed 11 Panday , D. , Panday , K. , Basnet , M. , Kafle , S. , Shah , B. ( 2017 ) Therapeutic drug monitoring of carbamazepine . International Journal of Neurorehabilitation , 4 , 245 . 12 MacKichan , J. , Duffner , P. , Cohen , M. ( 1981 ) Salivary concentrations and plasma protein binding of carbamazepine and carbamazepine 10, 11‐epoxide in epileptic patients . British Journal of Clinical Pharmacology , 12 , 31 – 37 . Google Scholar Crossref Search ADS PubMed 13 Shokry , E. , Villanelli , F. , Malvagia , S. , Rosati , A. , Forni , G. , Funghini , S. , et al. . ( 2015 ) Therapeutic drug monitoring of carbamazepine and its metabolite in children from dried blood spots using liquid chromatography and tandem mass spectrometry . Journal of Pharmaceutical and Biomedical Analysis , 109 , 164 – 170 . Google Scholar Crossref Search ADS PubMed 14 Zheng , N. , Zeng , J. , Ji , Q.C. , Angeles , A. , Aubry , A.-F. , Basdeo , S. , et al. . ( 2016 ) Bioanalysis of dried saliva spot (DSS) samples using detergent-assisted sample extraction with UHPLC-MS/MS detection . Analytica Chimica Acta , 934 , 170 – 179 . Google Scholar Crossref Search ADS PubMed 15 Zimmer , J.S. , Christianson , C.D. , Johnson , C.J. , Needham , S.R. ( 2013 ) Recent advances in the bioanalytical applications of dried matrix spotting for the analysis of drugs and their metabolites . Bioanalysis , 5 , 2581 – 2588 . Google Scholar Crossref Search ADS PubMed 16 Abdel‐Rehim , A. , Abdel‐Rehim , M. ( 2014 ) Dried saliva spot as a sampling technique for saliva samples . Biomedical Chromatography , 28 , 875 – 877 . Google Scholar Crossref Search ADS PubMed 17 Gallardo , E. , Barroso , M. , Queiroz , J. ( 2009 ) Current technologies and considerations for drug bioanalysis in oral fluid . Bioanalysis , 1 , 637 – 667 . Google Scholar Crossref Search ADS PubMed 18 Gallardo , E. , Queiroz , J. ( 2008 ) The role of alternative specimens in toxicological analysis . Biomedical Chromatography , 22 , 795 – 821 . Google Scholar Crossref Search ADS PubMed 19 Contin , M. , Mohamed , S. , Candela , C. , Albani , F. , Riva , R. , Baruzzi , A. ( 2010 ) Simultaneous HPLC–UV analysis of rufinamide, zonisamide, lamotrigine, oxcarbazepine monohydroxy derivative and felbamate in deproteinized plasma of patients with epilepsy . Journal of Chromatography B , 878 , 461 – 465 . Google Scholar Crossref Search ADS 20 Heideloff , C. , Bunch , D.R. , Wang , S. ( 2010 ) A novel HPLC method for quantification of 10 antiepileptic drugs or metabolites in serum/plasma using a monolithic column . Therapeutic Drug Monitoring , 32 , 102 – 106 . Google Scholar Crossref Search ADS PubMed 21 Patil , K. , Bodhankar , S. ( 2005 ) Simultaneous determination of lamotrigine, phenobarbitone, carbamazepine and phenytoin in human serum by high-performance liquid chromatography . Journal of Pharmaceutical and Biomedical Analysis , 39 , 181 – 186 . Google Scholar Crossref Search ADS PubMed 22 Pucci , V. , Raggi , M.A. ( 2005 ) Analysis of antiepileptic drugs in biological fluids by means of electrokinetic chromatography . Electrophoresis , 26 , 767 – 782 . Google Scholar Crossref Search ADS PubMed 23 Santagati , N.A. , Gotti , R. , Ronsisvalle , G. ( 2005 ) Simultaneous determination of phenytoin and dextromethorphan in urine by solid‐phase extraction and HPLC–DAD . Journal of Separation Science , 28 , 1157 – 1162 . Google Scholar Crossref Search ADS PubMed 24 Lanças , F.M. , Sozza , M.A. , Queiroz , M.E. ( 2003 ) Simultaneous plasma lamotrigine analysis with carbamazepine, carbamazepine 10, 11 epoxide, primidone, phenytoin, phenobarbital, and PEMA by micellar electrokinetic capillary chromatography (MECC) . Journal of Analytical Toxicology , 27 , 304 – 308 . Google Scholar Crossref Search ADS PubMed 25 Chollet , D.F. ( 2002 ) Determination of antiepileptic drugs in biological material . Journal of Chromatography B , 767 , 191 – 233 . Google Scholar Crossref Search ADS 26 Thormann , W. , Theurillat , R. , Wind , M. , Kuldvee , R. ( 2001 ) Therapeutic drug monitoring of antiepileptics by capillary electrophoresis: characterization of assays via analysis of quality control sera containing 14 analytes . Journal of Chromatography. A , 924 , 429 – 437 . Google Scholar Crossref Search ADS PubMed 27 ( 2018 ) Guidance for Industry : Bioanalytical Method Validation—2018 Version. Food and Drug Administration. https://www.fda.gov/downloads/drugs/guidances/ucm070107.Pdf (accessed January 3, 2018). 28 Scientific Working Group for Forensic Toxicology . ( 2013 ) Standard practices for method validation in forensic toxicology . Journal of Analytical Toxicology , 37 , 452 – 474 . Crossref Search ADS PubMed 29 Timmerman , P. , White , S. , Globig , S. , Lüdtke , S. , Brunet , L. , Smeraglia , J. ( 2011 ) EBF recommendation on the validation of bioanalytical methods for dried blood spots . Bioanalysis , 3 , 1567 – 1575 . Google Scholar Crossref Search ADS PubMed 30 Wilhelm , A.J. , den Burger , J.C.G. , Swart , E.L. ( 2014 ) Therapeutic drug monitoring by dried blood spot: progress to date and future directions . Clinical Pharmacokinetics , 53 , 961 – 973 . Google Scholar Crossref Search ADS PubMed 31 Luckwell , J. , Danielsson , Å. , Johnson , B. , Clegg , S. , Green , M. , Pierce , A. Applications and chemistry of cellulose papers for dried blood spots. In: Wenkui , L. , Mike , S.L. (eds). Dried Blood Spots: Applications and Techniques , 1st edition . John Wiley & Sons, Inc : Hoboken, New Jersey , 2014 ; pp 332 – 343 . Chapter 26. 32 Abu-Rabie , P. , Denniff , P. , Spooner , N. , Brynjolffssen , J. , Galluzzo , P. , Sanders , G. ( 2011 ) Method of applying internal standard to dried matrix spot samples for use in quantitative bioanalysis . Analytical Chemistry , 83 , 8779 – 8786 . Google Scholar Crossref Search ADS PubMed 33 Déglon , J. , Thomas , A. , Mangin , P. , Staub , C. ( 2012 ) Direct analysis of dried blood spots coupled with mass spectrometry: concepts and biomedical applications . Analytical and Bioanalytical Chemistry , 402 , 2485 – 2498 . Google Scholar Crossref Search ADS PubMed 34 Li , W. , Tse , F.L. ( 2010 ) Dried blood spot sampling in combination with LC–MS/MS for quantitative analysis of small molecules . Biomedical Chromatography , 24 , 49 – 65 . Google Scholar Crossref Search ADS PubMed 35 Déglon , J. , Versace , F. , Lauer , E. , Widmer , C. , Mangin , P. , Thomas , A. , et al. . ( 2012 ) Rapid LC–MS/MS quantification of the major benzodiazepines and their metabolites on dried blood spots using a simple and cost-effective sample pretreatment . Bioanalysis , 4 , 1337 – 1350 . Google Scholar Crossref Search ADS PubMed 36 Rani , S. , Malik , A.K. , Singh , B. ( 2012 ) Novel micro‐extraction by packed sorbent procedure for the liquid chromatographic analysis of antiepileptic drugs in human plasma and urine . Journal of Separation Science , 35 , 359 – 366 . Google Scholar Crossref Search ADS PubMed 37 Otero-Fernández , M. , Cocho , J.Á. , Tabernero , M.J. , Bermejo , A.M. , Bermejo-Barrera , P. , Moreda-Piñeiro , A. ( 2013 ) Direct tandem mass spectrometry for the simultaneous assay of opioids, cocaine and metabolites in dried urine spots . Analytica Chimica Acta , 784 , 25 – 32 . Google Scholar Crossref Search ADS PubMed 38 Edelbroek , P.M. , van der Heijden , J. , Stolk , L.M. ( 2009 ) Dried blood spot methods in therapeutic drug monitoring: methods, assays, and pitfalls . Therapeutic Drug Monitoring , 31 , 327 – 336 . Google Scholar Crossref Search ADS PubMed 39 Mei , J.V. , Alexander , J.R. , Adam , B.W. , Hannon , W.H. ( 2001 ) Use of filter paper for the collection and analysis of human whole blood specimens . The Journal of Nutrition , 131 , 1631S – 1636S . Google Scholar Crossref Search ADS PubMed 40 Boy , R.G. , Henseler , J. , Mattern , R. , Skopp , G. ( 2008 ) Determination of morphine and 6-acetylmorphine in blood with use of dried blood spots . Therapeutic Drug Monitoring , 30 , 733 – 739 . Google Scholar Crossref Search ADS PubMed 41 Peters , F.T. , Drummer , O.H. , Musshoff , F. ( 2007 ) Validation of new methods . Forensic Science International , 165 , 216 – 224 . Google Scholar Crossref Search ADS PubMed 42 Shah , V.P. , Midha , K.K. , Findlay , J.W. , Hill , H.M. , Hulse , J.D. , et al. . ( 2000 ) Bioanalytical method validation—a revisit with a decade of progress . Pharmaceutical Research , 17 , 1551 – 1557 . Google Scholar Crossref Search ADS PubMed 43 ( 2005 ) Validation of analytical procedures: text and methodology . Q2 (R1)—2005 Version. International Conference on Harmonization. https://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Quality/Q2_R1/Step4/Q2_R1__Guideline.pdf (accessed January 3, 2018). 44 Herkes , G.K. , McKinnon , G.E. , Eadie , M.J. ( 1989 ) Simultaneous quantitation of salivary carbamazepine, carbamazepine-10, 11-epoxide, phenytoin and phenobarbitone by high-performance liquid chromatography . Journal of Chromatography B: Biomedical Sciences and Applications , 496 , 147 – 154 . Google Scholar Crossref Search ADS 45 Đorđević , S. , Kilibarda , V. , Stojanović , T. ( 2009 ) Determination of carbamazepine in serum and saliva samples by high performance liquid chromatography with ultraviolet detection . Vojnosanitetski Pregled , 66 , 347 – 352 . Google Scholar Crossref Search ADS PubMed 46 Rukhadze , M.D. , Tsagareli , S.K. , Sidamonidze , N.S. , Meyer , V.R. ( 2000 ) Cloud-point extraction for the determination of the free fraction of antiepileptic drugs in blood plasma and saliva . Analytical Biochemistry , 287 , 279 – 283 . Google Scholar Crossref Search ADS PubMed 47 Hartley , R. , Lucock , M. , Cookman , J. , Becker , M. , Smith , I. , Smithells , R. , et al. . ( 1986 ) High-performance liquid chromatographic determination of carbamazepine and carbamazepine 10, 11-epoxide in plasma and saliva following solid-phase sample extraction . Journal of Chromatography B: Biomedical Sciences and Applications , 380 , 347 – 356 . Google Scholar Crossref Search ADS 48 MacKichan , J.J. ( 1980 ) Simultaneous liquid chromatographic analysis for carbamazepine and carbamazepine 10, 11-epoxide in plasma and saliva by use of double internal standardization . Journal of Chromatography , 181 , 373 – 383 . Google Scholar Crossref Search ADS PubMed 49 Langel , K. , Gunnar , T. , Ariniemi , K. , Rajamäki , O. , Lillsunde , P. ( 2011 ) A validated method for the detection and quantitation of 50 drugs of abuse and medicinal drugs in oral fluid by gas chromatography–mass spectrometry . Journal of Chromatography B , 879 , 859 – 870 . Google Scholar Crossref Search ADS 50 Ruiz , M.E. , Conforti , P. , Fagiolino , P. , Volonté , M.G. ( 2010 ) The use of saliva as a biological fluid in relative bioavailability studies: comparison and correlation with plasma results . Biopharmaceutics & Drug Disposition , 31 , 476 – 485 . Google Scholar Crossref Search ADS PubMed 51 Hartmann , C. , Smeyers-Verbeke , J. , Massart , D. , McDowall , R. ( 1998 ) Validation of bioanalytical chromatographic methods . Journal of Pharmaceutical and Biomedical Analysis , 17 , 193 – 218 . Google Scholar Crossref Search ADS PubMed 52 Dadgar , D. , Burnett , P.E. , Gerry Choc , M. , Gallicano , K. , Hooper , J.W. ( 1995 ) Application issues in bioanalytical method validation, sample analysis and data reporting . Journal of Pharmaceutical and Biomedical Analysis , 13 , 89 – 97 . Google Scholar Crossref Search ADS PubMed 53 Goldsmith , R.F. , Ouvrier , R.A. ( 1981 ) Salivary anticonvulsant levels in children: a comparison of methods . Therapeutic Drug Monitoring , 3 , 151 – 158 . Google Scholar Crossref Search ADS PubMed 54 Bansal , S. , DeStefano , A. ( 2007 ) Key elements of bioanalytical method validation for small molecules . The AAPS Journal , 9 , E109 – E114 . Google Scholar Crossref Search ADS PubMed 55 McDade , T.W. , Williams , S. , Snodgrass , J.J. ( 2007 ) What a drop can do: dried blood spots as a minimally invasive method for integrating biomarkers into population-based research . Demography , 44 , 899 – 925 . Google Scholar Crossref Search ADS PubMed 56 Shah , V.P. , Midha , K.K. , Dighe , S. , McGilveray , I.J. , Skelly , J.P. , Yacobi , A. , et al. . ( 1991 ) Analytical methods validation: bioavailability, bioequivalence and pharmacokinetic studies. Conference report . European Journal of Drug Metabolism and Pharmacokinetics , 16 , 249 – 255 . Google Scholar Crossref Search ADS PubMed © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For Permissions, please email: [email protected] This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)