Recovery and Stability of D 9 -Tetrahydrocannabinol Using the Oral-Eze w Oral Fluid Collection System and Intercept w Oral Specimen Collection Device

Transcription

1 Journal of Analytical Toxicology 2015;39: doi: /jat/bkv093 Special Issue Recovery and Stability of D 9 -Tetrahydrocannabinol Using the Oral-Eze w Oral Fluid Collection System and Intercept w Oral Specimen Collection Device Kimberly L. Samano 1, Lakshmi Anne 2, Ted Johnson 3, Kenneth Tang 2 and R.H. Barry Sample 1 * 1 Quest Diagnostics Incorporated, Employer Solutions, Renner Boulevard, Lenexa, KS 66219, USA, 2 Thermo Fisher Scientific, Clinical Diagnostics Division, Fremont, CA 94538, USA, and 3 Quest Diagnostics Incorporated, Employer Solutions, West Hills, CA 91304, USA *Author to whom correspondence should be addressed. barry.x.sample@questdiagnostics.com Oral fluid (OF) is increasingly used for clinical, forensic and workplace drug testing as an alternative to urine. Uncertainties surrounding OF collection device performance, drug stability and testing reproducibility may be partially responsible for delays in the implementation of OF testing in regulated drug testing programs. Stability of D 9 -tetrahydrocannabinol (THC) fortified and authentic specimens was examined after routine collection, transport and laboratory testing. Acceptable recovery and stability were observed when THC-fortified OF (1.5 and 4.5 ng/ml) was applied to Oral-Eze devices. Neat OF samples collected with Oral-Eze, processed per the package insert, and fortified with THC (3 and 6 ng/ml) were stable (+20%) at room temperature (21 258C), refrigerated (2 88C) and frozen (225 to 2158C) conditions up to 1 month, while samples collected with Intercept devices showed decreases at refrigerated and room temperatures. After long-term refrigerated or frozen storage, maximum reductions in THC concentrations were 42% for Oral-Eze and 69% for Intercept. After 1 year frozen storage, 80.7% of laboratory specimens positive for THC (3 ng/ml cut-off) by GC MS were reconfirmed positive (within 25%), with an average THC decrease of 4.2%. Specimens (n 5 47) processed with Oral-Eze (diluted) and tested via enzyme immunoassay were concordant with LC MS-MS results and showed 100% sensitivity and 95% specificity. Paired specimens collected with Oral-Eze and Intercept exhibited 98% overall agreement between the immunoassay test systems. Collectively, these data demonstrate consistent and reproducible recovery and stability of THC in OF after collection, transport and laboratory testing using the Oral-Eze OF Collection System. Introduction Oral fluid (OF) is an alternative matrix used for drugs of abuse (DoA) testing in forensic, clinical, workplace, criminal justice, driving under the influence of drugs (DUID) and drug treatment settings (1). As compared with traditional urine and blood matrices, OF collection is quick, straightforward and minimally invasive. Moreover, it is easily observed which makes adulteration or substitution difficult. Since OF is composed of secretions from salivary glands as well as interstitial fluid originating partially from the system circulation, parent drug as well as metabolite detection is possible (1, 2). Therefore, OF may serve to capture more recent drug use and/or impairment (1 4). Collection of OF may be achieved by pooling saliva and spitting (neat) or, more commonly today, with collection devices that are placed into the mouth. A subset of the data included in this manuscript were presented during a poster session at The International Association of Forensic Toxicologists (TIAFT) annual meeting in Buenos Aires, Argentina in The choice of specimen collection device and buffer preservative are important considerations which impact drug stability and analytical results (5 8). The Oral-Eze Oral Fluid Collection System (Oral-Eze) for DoA testing contains a collector with an absorbent pad incorporating a blue dye to indicate when an adequate volume of OF has been obtained and a transport tube containing 2 ml of a buffer preservative solution (BPS), which provides a 3-fold dilution of the specimen. Some commercially available collection devices do not standardize the specimen volume obtained which may compromise the sensitivity, accuracy and reproducibility of OF testing. Marijuana (Cannabis sativa) continues to be the most widely used illicit drug in the United States, and recent data from the Department of Health and Human Services National Survey of Drug Use and Health indicated that more than 22% of selfrespondents aged 12 years or older have used marijuana within the last month (9). D 9 -tetrahydrocannabinol (THC) is the main psychoactive constituent of marijuana and the predominant analyte detected in OF after smoking (10). To date, there is limited information available regarding the stability of THC in OF. Studies have been published using fortified, synthetic OF, or authentic OF collected with various collection devices (10 12). In particular, few studies have used real world specimens collected, transported, tested and stored under routine laboratory drug testing conditions. The aim of the present work is to evaluate the postcollection recovery and stability of THC with the Oral-Eze System using fortified and authentic OF specimens and to compare analytical results of authentic donor specimens collected with Oral-Eze and Intercept Oral Fluid Drug Test (Intercept) collection devices. Materials and methods Oral-Eze and Intercept collection devices Both Oral-Eze and Intercept are FDA-cleared devices for DoA drug testing in OF. Enzyme immunoassay and confirmation (GC MS or LC MS-MS) results are reported as positive or negative relative to a cut-off value of 3 ng/ml in neat OF. Drug concentrations in diluted (buffer/preservative) OF, whether collected by Oral-Eze or Intercept, were multiplied by three, to correlate results from initial or confirmation testing back to concentrations in neat OF collected. Unless otherwise specified, the concentrations reported for Oral-Eze and Intercept results refer to THC concentrations in neat OF, although specimens were tested as diluted OF samples. At least 10 min prior to OF collection, donors are instructed to refrain from eating, drinking, smoking, etc. OF is obtained by placing either device in the mouth, between the lower gum and cheek. The Oral-Eze System, but not the Intercept device, # The Author Published by Oxford University Press. All rights reserved. For Permissions, please journals.permissions@oup.com

2 has a built-in sample adequacy window to inform the test administrator when the expected volume of OF has been collected. A blue non-toxic indicator dye (food-coloring) migrates to the window as OF is adsorbed onto the cotton pad. When 1 ml of OF has been collected, the dye appears in the window signaling the collection is complete. After collection, the pad is ejected from the handle of the device into the collection tube which contains 2 ml of preservative buffer providing a 3-fold dilution of neat OF. The Intercept pad is treated with salts and gelatin to promote OF production during collection. Donors are instructed to swab the wand in their mouth until the pad is moist and keep the device in their mouth for a target collection time of 2 min (2 5 min), resulting in a nominal collection of 0.4 ml of OF. After collection, the wand is inserted into the oral specimen vial containing 0.8 ml of a BPS, providing a nominal 3-fold dilution of neat OF. The wand is scored such that the pad can be snapped off from the wand while inside the vial. THC stability in fortified and authentic DoA OF samples The data presented herein were obtained from a collaborative study across laboratories that routinely test for DoA. Initial testing was performed by enzyme immunoassay, and confirmation testing was achieved using both gas- and liquid-chromatography coupled to mass spectrometry. Whether initial or confirmatory, all analytical methods were validated per laboratory standard operating procedures and following quality control standards, prior to experimental testing. Fortified Study 1 was performed to investigate the performance of the Oral-Eze system using a pool of THC-free neat OF that was fortified with THC. The fortified samples were then applied onto the Oral-Eze collection device pad (n ¼ 2/level) until the blue indicator appeared. Neat OF equivalent target concentrations of 1.5 ng/ml (0.5) and 4.5 ng/ml (1.5) were chosen in this study to assess possible drug degradation around (+50%) the cut-off of 3 ng/ml. The pads were then transferred into the collection tubes containing BPS according to package insert instructions. As a result, the samples were diluted 3-fold, resulting in a final THC drug target concentration of 0.5 and 1.5 ng/ml. All samples were tested using a validated LC MS-MS method on the same day (Day 0) as application to the Oral-Eze collection device and were processed as routine specimens. The remaining specimen volume was stored at refrigerated (2 88C) and room (21 258C) temperatures with reanalysis at 7, 14 and 21 days to determine THC stability using a practical time course for OF collection and handling. Stability was considered acceptable if the mean recovery (n ¼ 2) of THC by LC MS-MS was within +20% of Day 0 control samples (2 88C). For Fortified Study 2, authentic drug-free OF was collected with Oral-Eze or Intercept devices, eluted per the manufacturers recommendations, pooled (separate Oral-Eze and Intercept pools) and then each pool fortified with THC at two different concentrations. The concentrations of 3 ng/ml (1) and 6 ng/ ml (2) were chosen to investigate the stability of processed (diluted in buffer/preservative) OF samples containing THC at or above the cut-off. Duplicate aliquots of each concentration were analyzed immediately (Day 0) by a validated GC MS method. The remainder of each devices pooled OF was divided into aliquots and then stored at C, 2 88C and frozen (225 to 158C) temperatures. Aliquots were stored for varying lengths of time at temperatures reflective of routine handling and drug testing with analysis conducted at 1, 3 and 4 weeks (all temperatures) and at 6 and 12 months (refrigerated and frozen temperatures). At each time point, 3 and 6 ng/ml aliquots were analyzed in duplicate (n ¼ 2). Data obtained from each duplicate set (n ¼ 2/concentration) were then normalized to Day 0 and combined (n ¼ 4 total observations). THC stability was considered acceptable at each time point if the mean drug concentration was within +20% of Day 0. Study 3 assessed the long-term stability of authentic donor specimens which were received at the laboratory for routine drug testing and subsequently screened and confirmed positive for THC via GC MS. After long-term frozen storage (225 to 2158C) for a minimum of 12 months, 143 random OF specimens were re-analyzed by the same validated confirmation method. Quantitative GC MS results from the initial confirmatory test were then compared with the re-analysis test result in order to assess long-term stability of THC in authentic OF after routine collection and storage using the Oral-Eze collection system. Method comparison samples Authentic donor OF samples were collected by spitting (neat) from individuals at a rehabilitation clinic (Donor Study 1 and 2). Informed consent was obtained from all participants for Quest Diagnostics Incorporated and Thermo Fisher Scientific to use biological specimens and data for technology development and research purposes. Donor Study 1 was conducted to evaluate the concordance between immunoassay (diluted OF) and LC MS-MS (neat and diluted OF) results. Neat OF samples were applied to the Oral-Eze pad and were processed to obtain the diluted (with BPS from the incorporated kit collection/transport tube) OF. The diluted OF (prepared diluted OF) samples were tested via enzyme immunoassay on a MGC 240 and by LC MS-MS as described below. Donor Study 2 was conducted to compare paired OF specimens collected with Oral-Eze and Intercept devices for the detection of THC in OF using both immunoassay (analyzed on an Olympus AU2700) and GC MS as outlined below. Oral-Eze and Intercept OF collection was performed sequentially for each device, with the order randomized evenly across the total sample collection. True positive (TP), false negative (FN), true negative (TN) and false positive (FP) designations were used to calculate sensitivity [TP/(TP/FN)] and specificity [TN/(TN þ FP)], where a positive is defined as a specimen containing THC at or above the cut-off (3 ng/ml, neat OF) and a negative is defined as a specimen containing no detectable drug or THC below the cut-off. Enzyme immunoassays OF specimens from Donor Study 1 and 2 were screened for THC on a Thermo Scientific MGC 240 benchtop analyzer (Thermo Fisher Scientific, Fremont CA) or Beckman AU2700 Automated Chemistry Analyzer (Beckman Coulter, Brea, CA), respectively, using the CEDIA w Cannabinoids OFT Assay (Thermo Fisher Scientific). The kit utilizes b-galactosidase recombinant DNA technology, which produces a colorimetric change in the presence of THC and is monitored at 570 and 660 nm dual wavelengths. Analytical runs were calibrated at the cut-off (1 ng/ml, diluted OF) on both systems using CEDIA THC OFT calibrators (Thermo THC Recovery and Stability with Oral-Eze w and Intercept w 649

3 Fisher Scientific), with negative (0 ng/ml) and high (10 ng/ml, diluted OF) calibrators included on the MGC240 analyzer. The assay results are reported as a positive or negative relative to the calibrator cut-off value of 3 ng/ml (1 ng/ml diluted). Therefore, a sample that exhibits a change in absorbance (DA) value equal to or greater than the value of the cut-off calibrator was considered positive, whereas a sample that exhibits a DA value lower than the calibrator was considered negative. GC MS THC and THC-D 3 internal standard (ISTD; Cerilliant Corporation) were added to each calibrator, QC and sample to achieve a final concentration of 100 ng/ml (equivalent concentration in neat OF). Two milliliters of hexane ethyl acetate (90:10, v/v) containing 3% acetic acid were added to each 300 ml aliquot of calibrator, QC and sample. The samples were rotator-mixed for 10 min and then centrifuged for 5 min at 2,500 g. The organic layer was removed and transferred to a clean tube. Eluates were evaporated to dryness and derivatized using N, O-bis (trimethylsilyl)trifluoroacetamide, or BSTFA containing 1% trimethylchlorosilane (TMCS), for 20 min at 708C. Extracts were injected onto a 6890N Agilent GC System equipped with Deans Switch and coupled to a 5973 Agilent Mass Selective Detector. Each confirmation run was calibrated with calibrators utilizing deuterated internal standards for retention time and quantification reference, at a target concentration of 1.5 ng/ml for THC. Negative, positive and threshold cut-off QC samples were analyzed with each analytical run. Quantification of THC in OF samples was achieved using a 3.0 ml injection ( pulsed splitless) onto analytical column 1 consisting of a DB-17 ms column (15 m, 0.25 mm, 0.25 mm; Agilent Technologies, Santa Clara, CA) with an inlet temperature of 2508C and a pulse pressure of 40.0 psi, and analytical column 2 consisting of a DB-1 ms column (15 m, 0.25 mm, 0.25 mm; Agilent Technologies) using ultra-pure helium carrier gas. Selected ion monitoring acquisition mode was used to monitor the following ions (quantification ions in bold): m/z 303, 371, 386 for THC and m/z 346, 374 for THC-D 3.GC/MSD Chemstation software (Agilent Technologies) was used for data acquisition, and extracts were considered acceptable if the negative control was less than the LOD, QC controls were within +25% of the established mean and peaks were symmetrical with valid ion ratios (+20%), and retention time (+2%). LC MS-MS Quantitative values for THC were obtained using a linear calibration model (origin included and weighted [1/X]) ranging from 0.5 to 20.0 ng/ml, with QC samples targeted at 1.5, 4.0, 8.0 and 16 ng/ml. To 200 ml of neat OF or matrix-matched OF calibrator and QCs, 20 mlofthc-d 3 ISTD was added to silianized glassware to achieve a final concentration of 40 ng/ml. Drug was isolated by liquid liquid extraction with 400 ml of acetic acid added to samples and mixed prior to the addition of 2.0 ml of hexane. The samples were mixed for 1 h and centrifuged for 15 min at 2,360 g. The organic layer was removed, transferred to a clean tube and evaporated to dryness. Eluates were reconstituted with 50 ml of mobile phase solution, comprised of methanol water (90:10 v/ v) containing 5 mm ammonium formate. Then, 30.0 ml extracts were injected onto a Waters 2795 Alliance HPLC system using a Nova Pak C18 reverse phase column (150 mm, 2.1 mm, 4 mm; Waters Corporation, Millford, MA) coupled to a Micromass w Quattro Micro triple quadrapole mass spectrometer equipped with a Z-Spray electrospray ionization source operating in positive mode. The column temperature was maintained at 508C with a constant flow rate of ml/min, and multiple reaction monitoring was used to monitor the following transition ions: m/z for THC and m/z for THC-D 3. Masslynx TM NT 4.0 and Quanalynx software (Waters Corporation) were used for data acquisition, system control and data processing. The lower limit of quantitation (LLOQ) was fixed at 0.5 ng/ml with QC controls considered acceptable if values were within +10% of the target value, with integrated peaks within +0.2 min of the retention time standard. A summary of the rationale, experimental design and analysis methods is shown in Table I. Results Recovery/stability experiments Results for Fortified Study 1 are presented in Table II. Recovery and stability of THC fortified into neat OF and applied to Oral-Eze devices were within +20% target concentration (range, %) after room temperature (21 258C) and refrigerated (2 88C) storage, measured at 7, 14 and 21 days. Stability of samples fortified with THC at 0.5 ng/ml (diluted) ranged from 91 to 104% under refrigerated conditions and 85 94% when maintained at room temperature across the 3-week period. Furthermore, samples fortified with 1.5 ng/ml (diluted) of THC remained stable with at least 94% recovery of THC across all time points after refrigerated and room temperature storage. The results for Fortified Study 2 are shown in Table III. OF collected with Oral-Eze devices then fortified with THC and stored at C, 2 88C and 225 to 2158C temperatures were within +20% of original THC concentrations for up to 1 month, with Table I Summary of Rationale, Experimental Design and Analysis Table Study Purpose Matrix [Neat OF] [Diluted OF] n Method II Fortified Study 1 THC recovery/stability from Oral-Eze device, multiple time THC-fortified blank OF applied to device 1.5 ng/ml 0.5 ng/ml 2 LC MS MS points/temperatures 4.5 ng/ml 1.5 ng/ml III Fortified Study 2 THC stability with Oral-Eze and Intercept devices, multiple time Blank OF collected with device, fortified 3ng/mL 1ng/mL 2 GC MS points/temperatures with THC 6ng/mL 3ng/mL IV Study 3 Long-term stability of authentic OF specimens collected with Oral-Eze and stored 12 months Authentic THCþ workplace drug testing specimens N/A Linear range 1 60 ng/ml 140 EIA GC MS V Donor Study 1 Screening/confirmation method concordance with specimens collected with Oral-Eze Authentic neat OF applied to device N/A N/A 47 EIA LC MS MS VI Donor Study 2 Comparison of Oral-Eze and Intercept devices for the detection of THC Paired authentic OF collected with devices N/A N/A 472 EIA GC MS 650 Samano et al.

4 Table II Recovery and Stability of THC Fortified in Oral Fluid and Applied to Oral-Eze Devices as Determined by LC MS-MS THC (ng/ml, diluted) diluted with BPS Day 0 Day 7 Day 14 Day C 2 88C C 2 88C C 2 88C C Level 1 a (0.5) Level 2 b (1.5) Level 1 as % Day Level 2 as % Day a Target concentration of 1.5 ng/ml in neat oral fluid (n ¼ 2). b Target concentration of 4.5 ng/ml in neat oral fluid (n ¼ 2). Table IV Postcollection Stability of THC (ng/ml, Neat) in Authentic OF Collected with Oral-Eze After Frozen Storage (225 to 158C) for 1 Year Group Concordance a n % Mean SD 1 +25% % % ULOL b 10 ND ND ND, not determined. a Comparison of initial and re-analysis testing by GC MS. b Upper limit of linearity (60 ng/ml). Table III Stability of THC Fortified in OF Collected with Oral-Eze and Intercept as Determined by GC MS a Storage temperature Average (% CV) THC recovery (%) b from Oral-Eze 1 Week 3 Weeks 1 Month 6 Months 12 Months C 110 (2.7) 107 (9.8) 111 (4.2) NT NT 2 88C 111 (0) 121 (2.2) 119 (1.5) 58 (5.4) 61 (9.7) 225 to 2158C 111 (0) 107 (4.9) 114 (3.4) 65 (10.7) 75 (8.9) Storage Temperature Average (%CV) THC recovery (%) b from Intercept 1 Week 3 Weeks 1 Month 6 Months 12 Months C 67 (5.9) 39 (18.6) 32 (15.5) NT NT 2 88C 89 (0.3) 74 (5.6) 79 (10.2) 68 (5.7) 31 (16.6) 225 to 2158C 98 (5.1) 89 (6.9) 82 (8.0) 99 (8.0) 83 (2.7) NT, not tested. a Target concentrations of 3 and 6 ng/ml in neat OF (n ¼ 4). b Recovery (%) normalized to Day 0. the exception of week 3 results at 2 88C (range, %). Fortified Oral-Eze aliquots stored at 2 88Cfor6or12monthsdisplayed an average decrease in OF THC concentrations of 41%. While, an average decrease in THC of 35% and 25% was observed with Oral-Eze aliquots stored frozen for 6 and 12 months, respectively. During this study, THC stability was more variable in OF collected with Intercept devices versus Oral-Eze, with coefficients of variation (% CV) ranging from 0.3 to 18.6% and 0 to 10.7% for each device, respectively. In particular, Intercept THC concentrations were not consistently stable (+20%) after room temperature and refrigerated storage. Table III reveals decreases in THC ranging from 11 68% observed in fortified Intercept aliquots stored from 1 week to 1 month at room temperature and refrigerated storage. OF collected with Intercept, fortified with THC and stored refrigerated were within +20% of original THC concentrations for 1 week, but not at the 3 week and 1 month time points. Decreases in THC of 32% and 69% were observed with THC-fortified Intercept aliquots after 6 and 12 months of refrigerated storage, respectively. In contrast, intercept THC concentrations after frozen storage were within +20% of the original THC concentrations across all time points, with decreases in THC ranging from 1 to 17%. Authentic specimens (n ¼ 143) initially confirmed positive for THC during routine laboratory testing were selected for reanalysis of THC after more than 12 months frozen storage at 225 to 2158C. However, data from three samples were not included due to insufficient specimen volume for re-analysis. Negative specimens (n ¼ 3) included in this study were reconfirmed negative for THC (data not shown). Specimen concentrations ranged from 1 ng/ml to greater than the assays upper limit of linearity (ULOL), 60 ng/ml. Table IV summarizes the decrease or Table V Concordance of Immunoassay (Diluted) and LC MS-MS (Diluted) Oral Fluid Donor Samples Processed with the Oral-Eze Oral Fluid Collection System CEDIA cannabinoids OFT a assay results LC MS-MS results Less than 50% below cut-off b Near cut-off negative c Near cut-off More than 50% positive d above cut-off e Negative Positive 0 1 f 3 22 a Oral fluid testing. b Concentration range,1.5 ng/ml. c Concentration range.1.5 and,3.0 ng/ml. d Concentration range.3.0 and,4.5 ng/ml. e Concentration range.4.5 ng/ml. f Concentration by LC MS-MS was 2.88 ng/ml. increase in THC concentration and standard deviation for THC-positive specimens quantified within the linear range (n ¼ 130) for the +25%, %, and greater than +50% groups. All 10 specimens initially reported above the THC assays ULOL were reconfirmed at concentrations greater than the ULOL. OF specimens (n ¼ 130) originally reported positive for THC by GC MS were reconfirmed for the presence of THC, with 91% of Group 1 within +20%, and an average decrease in THC of 4.2%. The decrease in THC concentration for Group 2 ranged from 26 to 50%, with an average decrease of 35.2%. The three specimens in Group 3 (greater than +50%) showed decreases in THC that ranged from 52.9 to 63.3%, with an average loss of 59.1%. Method comparison results Analysis of Oral-Eze processed OF specimens from Donor Study 1 show 98% overall agreement (100% sensitivity, 95% specificity) between immunoassay and LC MS-MS (3.0 ng/ml cut-off) results, shown in Table V. The one discordant specimen had a concentration of 2.88 ng/ml (0.12 ng/ml below cut-off). In Donor Study 2, concentrations of THC from paired OF specimens (Oral-Eze and Intercept devices) were analyzed by enzyme immunoassay, and results are shown in Table VI. The sensitivity and specificity between Oral-Eze and Intercept devices were 89 and 99%, respectively. The overall agreement between the two immunoassay collection and testing systems was 98%. Discussion Analysis of DoA in OF presents its own specific set of advantages and challenges, which differ from that of urine drug testing (see THC Recovery and Stability with Oral-Eze w and Intercept w 651

5 Table VI Overall Accuracy Between Intercept and Oral-Eze Immunoassay Results a Oral-Eze results Intercept results Overall accuracy Negative Positive Negative Sensitivity 89% Positive 4 33 Specificity 99% Overall agreement 98% a Near cut-off ( borderline ) samples included. review Bosker et al. (1)). In contrast to many other commonly abused drugs, the lipophilic nature of THC makes recovery from specimen collection devices a unique concern (1, 13, 14). Previously, a series of in vitro drug recovery studies were performed using fortified OF at physiologically relevant concentrations (1 25 ng/ml) for both THC and THC-COOH (13, 14). In the study, THC extraction from OF devices was poor, with an estimated recovery of 25% THC from ORALscreen w, Salivette w, Hooded Collector, Finger Collector w and Intercept w devices. The authors further concluded that THC-COOH recovery was otherwise unacceptable by analytical standards as Intercept demonstrated the highest recovery of THC-COOH at 53% (13, 14). A separate study showed the recovery of THC was variable under realistic transport and laboratory conditions using Quantisal TM (60 85%), Certus w (10 80%) and StatSure (62 106%) OF collectors, targeted at both low (15 ng/ml) and high (240 ng/ml) concentrations (8). On the other hand, 82% THC (8 ng/ml) extraction efficiency has been reported using the Quantisal device with a coefficient of variation calculated at 5.2% (11). Furthermore, Quintela et al. reported improved recovery ( %) of THC at 2, 4, and 8 ng/ml with Quantisal (15). A comprehensive evaluation of 10 OF collectors indicated the greatest recovery of THC of 85.5% from StatSure, with the least THC recovery of 37.6% obtained from Intercept; however, the concentration of the THC (1,000 ng/ml) fortified samples in this study were significantly higher than previous studies (5) and concentrations used for the experiments presented here. Because discrepant recovery results have been documented in the literature across several OF collection devices using a range of target concentrations, we wanted to investigate THC recovery using Oral-Eze. As part of a collaborative effort, quantitative values were obtained by mass spectrometric techniques, with separation achieved using both gas- and liquid-chromatography as described previously. It should be noted that the effect(s) of using variable analytical methods were not evaluated in this study (e.g., biases); however, each method was validated prior to implementation. Additionally, laboratories routinely quantify THC in biological matrices using GC MS or LC MS methodologies; therefore, we would not expect this to alter the results and conclusions presented in this article. Aliquots targeted at 0.5 and 1.5 ng/ml of THC in diluted OF showed acceptable recovery and were within +20% (85 104%) of fortified concentrations (Table II). Compared with other OF collection devices, our data demonstrate that acceptable THC recovery and stability are observed with Oral-Eze collection devices when maintained for up to 21 days at room temperature (21 258C) or refrigerated (2 88C) conditions with reliable and reproducible results. After transportation and initial testing in the laboratory, drug-positive OF often requires long-term storage prior to its final disposition. As a result, studies were conducted to evaluate the post-collection stability of THC with Oral-Eze using both fortified and authentic OF specimens at room temperature, refrigerated and frozen conditions. The study results (Table III) indicate that OF samples diluted with Oral-Eze BPS and fortified with THC (3 and 6 ng/ml) are stable ( %) at room temperature (21 258C), refrigerated (2 88C) and frozen (225 to 2158C) conditions up to 1 month. These results are consistent with previously described THC stability of 80% (22 288C) and 90% (2 88C) using Quantisal pads fortified with 8 ng/ml of THC (11). When Intercept was evaluated, our study revealed that THC recovery was variable after room temperature (32 67%) and refrigerated storage (31 89%), although acceptable stability was observed after frozen storage across time points (Table III). Crouch et al. reported similar findings, with THC stabilities of % (48C), % (218C) and 72 87% (225 to 2158C) after 2, 3 and 6 weeks (13). In contrast, one study reported marginal stability (+25%) using pooled real OF samples after collection with the Intercept device (16). Relative to THC concentrations at T 0 (0.6 and 1.5 ng/ml), Lund et al. showed concentrations of THC ranged from 76 to 95% (room temperature) and 74 to 83% (refrigerated) after 1 week. In the same study, THC concentrations were % of original concentrations after 3 or 11 months frozen storage (16). These differences in stability after room temperature and refrigerated conditions may be explained by the different lengths of time used between studies, as the greatest THC loss (67%) in this study was observed 4 weeks after room temperature storage. When stored frozen (225 to 2158C), OF collected with Intercept and fortified with THC showed acceptable stability of % after short-term (1 4 weeks) storage and % (6 12 months) after long-term storage, which has been documented previously (5, 13, 16). Oral-Eze OF samples fortified with THC exhibited up to a 42% reduction in THC after 6 and 12 months frozen storage (Tables II and III), which may be reflective of the nature of the specimen preparation and sample size (n ¼ 2/concentration/time point/temperature). As a complementary and more representative approach, the stability of THC was determined in THC-positive (n ¼ 140) routine workplace drug testing specimens (Table IV). After at least 1 year frozen storage, 80.7% (105/130) of the specimens had THC concentrations within +25% of the original quantitative values. Remarkably, these specimens showed an average decrease in THC concentration of only 4.2% one year after initial analysis, with concentrations distributed across the assays linear range (1 60 ng/ml). Additionally, during the reanalysis, OF specimens previously reported negative were reconfirmed negative. While a small subset of the sample population (n ¼ 3) showed larger differences, these data suggest that spurious THC results are not generated after long-term storage. Further, only 2% (3/143) of stored authentic laboratory specimens were unable to be reanalyzed due to insufficient specimen volume, which suggests that Oral-Eze oral fluid collection and recovery volume are adequate for routine drug testing. Even though aqueous in nature, unlike urine, OF contains salivary enzymes and other proteins which may affect the stability and recovery of drug in an OF specimen. As a result, buffers and/or preservative solutions are incorporated into OF collection systems, resulting in an up-front dilution of neat OF. To examine this variable using the Oral-Eze collection system, fortified and authentic neat OF specimens were each applied to an Oral-Eze collection 652 Samano et al.

6 pad and routinely processed, resulting in BPS diluted OF specimens that were tested by immunoassay and LC MS-MS. Data in Table V show 98% overall agreement of immunoassay and LC MS-MS results in authentic specimens (Donor Study 1) and data in Table II show recovery of concentrations within 15% in the fortified specimens. Lastly, initial and confirmation results were compared using OF specimens ( paired Oral-Eze and Intercept OF collection) obtained from donors at a rehabilitation clinic (Donor Study 2). Sensitivity was 89% and specificity was 99%, with an overall agreement of 98% between Intercept and Oral-Eze specimens analyzed by enzyme immunoassay (Table VI). Furthermore, data analysis from Donor Study 2 indicates 3-fold higher average THC concentrations (diluted) in the Oral-Eze collected specimens as compared to Intercept collected specimens. Quantitative GC MS analysis of these paired specimens revealed average THC concentrations (diluted) of 17 ng/ml (167% CV) for Intercept and 51 ng/ml (144% CV) for Oral-Eze collected specimens. THC concentration ranges for Intercept were ng/ml, with Oral-Eze ranges of ng/ml observed. Data from the Drug Testing Index TM indicate that THC positivity in OF workplace drug testing has steadily risen from 2011 to 2014, which coincides with the implementation of Oral-Eze (17). Whether the differences in THC positivity rates are a result of differences in OF collection device, volume, dilution, drug recovery, stability or a combination of these or other variables is still unknown. A recent study by Desrosiers et al. reported no significant differences in THC concentrations between Oral-Eze and StatSure collection devices but reported significantly higher THC-COOH detection and improved stability of cannabinol (CBN) with Oral-Eze (18). Similar findings (19) were reported for THC as well as for the additional cannabinoids, THC-COOH and cannabidiol (CBD), in authentic OF samples following controlled cannabis administration. Concentrations were stable in % (THC), 78 89% (CBD) and % (THC-COOH) of samples from controlled cannabis experiments collected using the Oral-Eze collection system after room temperature and refrigerated storage for 1 and 4 weeks (19). OF data collected from occasional and chronic cannabis smokers using Oral-Eze devices indicate that THC-COOH concentrations are significantly higher in frequent smokers (4/week) versus occasional smokers (,2/week) when tested up to 30 h after cannabis smoking (20). Median THC concentrations for the last positive sample were 1.1 mg/l ( ) in frequent smokers and 0.9 mg/l ( ) in occasional smokers, with a T max of 0.5 and 0.8 (0.5 2) h, respectively. In comparison, the median THC-COOH concentrations for the last positive sample were 64.9 ng/l ( ) in frequent smokers and 21.2 ng/l ( ) in occasional smokers, with a T max of 2 (0.5 30) and 2 (0.5 3) h, respectively (20). Identification and quantification of THC-COOH may serve as a marker for cannabis use in addition to, or in the absence of THC, as concentrations have been shown to rapidly decrease in the first 1 2 h after smoking (21). Our data show that THC is stable in OF collected with Oral-Eze; this is supported by independent studies which demonstrate THC, THC-COOH, CBN and CBD detection in OF following controlled cannabis smoking experiments (18 20). Conclusions Accurate and precise analysis of DoA in OF depends not only on the collection device, transport tube and accompanying buffer system, but on initial and confirmatory testing methods. Therefore, a careful evaluation of each component individually and in combination with the overall testing methodology is necessary prior to implementation of an OF drug testing program. This article evaluated the recovery, stability and analytical method performance for the identification and quantification of THC in OF processed by either of two FDA-cleared OF collection devices. Collectively, the data demonstrate that the Oral-Eze Oral Fluid Collection System provides consistent and reproducible recovery and stability of THC from the point of collection through laboratory testing, storage and final disposition. Acknowledgments The authors thank Thuy Pham and Vani Bodepudi for their technical assistance in the study, as well as Carol A. Thompson for her assistance during the preparation of the manuscript. This work was supported by both Thermo Fisher Scientific and Quest Diagnostics Incorporated. Oral-Eze is the registered trademark of Quest Diagnostics Incorporated. T.J., K.L.S and R.H.B.S are employees of Quest Diagnostics, and L.A., K.T., are employees of Thermo Fisher. No external financial support was provided for this study. References 1. Bosker, W.M., Huestis, M.A. (2009) Oral fluid testing for drugs of abuse. Clinical Chemistry, 55, Aps, J.K.M., Martens, L.C. (2005) Review: the physiology of saliva and transfer of drugs into saliva. Forensic Science International, 150, Niedbala, R.S., Kardos, K.W., Fritch, D.F., Kardos, S., Fries, T., Waga, J. (2001) Detection of marijuana use by oral fluid and urine analysis following single dose administration of smoked and oral marijuana. Journal of Analytical Toxicology, 25, Schramm, Q., Smith, R.H., Craig, P.A. (1992) Drugs of abuse in saliva: a review. Journal of Analytical Toxicology, 16, Langel, K., Engblom, C., Pehrsson, A., Gunnar, T., Ariniemi, K., Lillesunde, P. (2008) Drug testing in oral fluid-evaluation of sample collection devices. Journal of Analytical Toxicology, 32, Ventura, M., Pichini, S., Ventura, R., Zuccaro, P., Pacifica, R., de la Torre, R. (2007) Stability studies of principal illicit drugs in oral fluid: preparation of reference materials for external quality assessment schemes. Therapeutic Drug Monitoring, 29, Ventura, M., Pichini, S., Ventura, R., Leal, S., Zuccaro, P., Pacifica, R. et al. (2009) Stability of drugs of abuse in oral fluid collection devices with purpose of external quality assessment schemes. Therapeutic Drug Monitoring, 31, Wille, S.M.R., DiFazio, V., del Mar Ramirez-Fernandez, M., Kummer, N., Samyn, N. (2013) Driving under the influence of cannabis: pitfalls, validation, and quality control of a UPLC MS-MS method for the quantification of tetrahydrocannabinol in oral fluid collected with StatSure, Quantisal, or Certus collector. Therapeutic Drug Monitoring, 35, SAMHSA, Center for Behavioral Health Statistics and Quality, National Survey on Drug Use and Health (NSDUH), Lee,D.,Milman,G.,Schwope,D.M.,Barnes,A.J.,Gorelick,D.A., Huestis, M.A. (2012) Cannabinoid stability in authentic oral fluid after controlled cannabis smoking. Clinical Chemistry, 58, Moore,C.,Vincent,M.,Rana,S.,Coulter,C.,Agrawal,A.,Soares, J.(2006) Stability of delta(9)-tetrahydrocannabinol (THC) in oral fluid using the Quantisal collection device. Forensic Science International, 164, THC Recovery and Stability with Oral-Eze w and Intercept w 653

7 12. Moore, C., Rana, S., Coulter, C. (2007) Simultaneous identification of 2-carboxy-tetrahydrocannabinol, tetrahydrocannabinol, cannabinol and cannabidiol in oral fluid. Journal of Chromatography B, 852, Crouch, D.J. (2005) Oral fluid collection: the neglected variable in oral fluid testing. Forensic Science International, 150, Crouch, D.J., Day, J., Baudys, J. (2004) Evaluation of saliva/oral fluid as an alternate drug testing specimen final report. National Institute of Standards and Technology (NISTR), gov/pdffiles1/nij/grants/ pdf. 15. Quintela, O., Crouch, D.J., Andrenyak, D.M. (2006) Recovery of drugs of abuse from the Immunalysis Quantisal oral fluid collection device. Journal of Analytical Toxicology, 30, Lund, H.M.E., Øiestad, E.L., Gjerde, H., Christophersen, A.S. (2011) Drugs of abuse in oral fluid collected by two different sample kits stability testing and validation using ultra performance tandem mass spectrometry analysis. Journal of Chromatography B, 879, Quest Diagnostics Drug Testing Index (DTI) 2015 release. (2015) drug-testing (date last accessed June 28, 2015). 18. Desrosiers, N.A., Milman, G., Mendu, D.R., Lee, D., Barnes, A.J., Gorelick, D.A. et al. (2014) Cannabinoids in oral fluid by on-site immunoassay and GC-MS using two different oral fluid collection devices. Analytical & Bioanalytical Chemistry, 406, Anizan, S., Bergamaschi, M.M., Barnes, A.J., Milman, G., Desrosiers, N., Lee, D. et al. (2014) Impact of oral fluid collection device on cannabinoid stability following smoked cannabis. Drug Testing and Analysis, 7, Newmeyer, M.N., Desrosiers, N.A., Lee, D., Mendu, D.R., Barnes, A.J., Gorelick, D.A. et al. (2014) Cannabinoid disposition in oral fluid after controlled cannabis smoking in frequent and occasional smokers. Drug Testing and Analysis, 10, Lee, D., Huestis, M.A. (2013) Current knowledge on cannabinoids in oral fluid. Drug Testing and Analysis, 6, Samano et al.