Journal of Analytical Toxicology, Vol. 30, November/December2006
Detection of THCA in Oral Fluid by GC-MS-MS
David Day, David J. Kuntz*, and Michael Feldman
LabOne, Inc., West Valley City, Utah 84120
Lance Presley
LabOne, Inc., Lenexa, Kansas 66219
I Abstract I
A major marijuana metabolite, 11-nor-9-carboxytetrahydrocannabinol (THCA), has been identified in oral
fluids from donors that previously tested positive for
Ag-tetrahydrocannabinol (THC). The method consisted of
solid-phase extraction of the oral fluid samples followed by gas
chromatography-tandem mass spectrometry analysis of the
extracts. Testing for THCA was performed on 223 oral fluid
samples previously analyzed for THC. The THCA assay was
linear from 10 to 240 pg/mL. The mean recovery of spiked
THCA in oral fluid was 104%, and precision was 4% at
20 pg/mL using fortified negative samples. This method was
rugged and robust, providing detection and quantification of
THCA in oral fluids at levels not previously reported. Results
of this study showed that THCA was detectable in 21 of 26 oral
fluid samples previously reported positive for THC. The range of
concentrations from these samples was from 10 pg/mL
up to 142 pg/mL THCA.
Introduction
Using oral fluid for the detection of drugs, including marijuana, has become popular for employment drug testing for
several reasons: 1. oral fluid collection is less intrusive than
urine collection; 2. this collection can be witnessed to minimize adulteration and sample switching or substitution; and 3.
no special collection facilities are required. The current oral
fluid analysis to detect marijuana use is to analyze for Ag_
tetrahydrocannabinol (THC). The notice of proposed rule
making (NPRM) by the Substance Abuse and Mental Health
Services Administration (SAMHSA)for alternate matrixes (Fed
Register Notice #FR Doc 04-7984) requires the simultaneous
collection of urine to be used for the detection of marijuana
use, by targeting the metabolite, 11-nor-9-carboxy-tetrahydrocannabinol (THCA). This is because of a perception that
testing for THC alone in oral fluid does not preclude possible
passive contamination. Using highly sensitive gas chromatog-
raphy-tandem mass spectrometry (GC-MS-MS) methodology,
this laboratory tested previously analyzed oral fluid samples to
determine if THCA could be detected. The purpose of this
study was to determine if THCA can be quantitatively measured in oral fluids from marijuana users and thereby eliminate the requirement for collection and analysis of urine
samples.
Urine testing for THCAis currently performed on a routine
basis. THCA is the major metabolite of THC and can be detected in urine from hours to days after marijuana is administered (1). High-volume methodologies have been developed
for detection of THCAin urine (2). Urine THCAlevels are used
to indicate marijuana use; however, urine is not the only biological matrix available for detection of THCA.
The most useful biological matrixes for detection of marijuana use are blood, hair, urine, sweat, and oral fluid. Mechanisms of mass transfer of THCA from blood to urine are well
understood. Kidney filtration is the most obvious of these
transfer routes. Additionally, transfer of THCA from the blood
to the hair has also been investigated and theorized (3). Although drug transport from blood to saliva is welt understood
(4,5), the mechanisms of THCA transfer from blood to saliva
require further investigation. Schramm et al. (6) purported to
identify THCA in a single saliva specimen using HPLC and
mass spectrometry.
In addition to abused drug detection, oral fluid testing is
used as a diagnostic tool for viral infections and monitoring
disease states (7). Moreover, oral fluid testing remains a useful
analytical tool in forensic toxicology, including for workplace
drug testing.
This article describes a useful oral fluid test for THCAat low
picogram-per-milliliter levels. Testing for THCA in oral fluid
samples, that were shown to contain THC, was performed on
authentic donor samples previously tested by two laboratories.
Materials and Methods
Oral fluid samples
* Author to whom correspondence should be addressed: David ). Kuntz, LabOne, Hayes Bldg.,
Unit C, 2282 S. Presidents Dr., West Valley City, UT 84120. E-mail: [email protected].
Oral fluid samples were collected from 223 subjects with the
Intercept| oral fluid collection device manufactured by Ora-
Reproduction(photocopying)of editorialcontentof thisjournal is prohibitedwithoutpublisher'spermission.
645
Journal of Analytical Toxicology, Vol. 30, November/December 2006
Sure Technologies, Inc. (Bethlehem, PA), and were screened
for THC using the manufacturer's oral fluid reagents following
instructions described in the package insert. The Intercept
device collects oral fluid onto a cotton pad and the pad is
transferred to a polypropylene tube with a buffer. Oral fluid
from drug-free volunteers was collected, pooled, and used as a
negative control. The negative oral fluid was certified free
from THC and THCAby GC-MS-MS. Standards and controls
were prepared in ethanol and diluted for testing into negative
oral fluid. THCA,THe, THe-d3, and THCA-d3were purchased
from Cerilliant (Austin, TX). Each standard material was independently verified to be 99.9% pure. 1,1,1,3,3,3-Hexafluoro2-propanol (HFIP) was purchased from AlfaAesar (Ward Hill,
MA), and pentafluoroproprionic anhydride (PFAA)was purchased from Pierce (Rockford,IL).
The oral fluid samples, obtained from two drug-testing laboratories, were separated into three groups. Samples in Group
A were screened negative by Enzyme Linked Immunoassay
(ELISA)for THe (cutoff t.0 ng/mL, cross-reactivitywith THCA
at 0.33 ng/mL) and consisted of 144 samples. Group B contained 53 samples that had screened positive for THe by
ELISA, but failed to confirm for THC using negative ion chemical ionization GC-MS-MS (cutoff 1.0 ng/mL, lower limit of
quantitation 0.4 ng/mL). Group C contained 26 samples that
screened positive for THe by ELBA and confirmed positive for
THe by GC-MS-MS (Table I). All samples in these groups
were tested for THCAby GC-MS-MS.
Table I. Between-Run Precision (23 Replicates)
Batch
Run #
Measured
Amount (pg/mL)
Measured
Amount (pg/mL)
1
2
3
4
5
10.00
9.70
11.90
10,80
8,50
56.9
52.1
51.1
59,3
62.4
6
7
8
9
t0
10.00
10.80
8.50
12.50
13.00
11
11,90
12
13
14
15
16
17
18
7.5O
9.70
9.90
9.00
10,80
10.80
8.40
55.4
56.6
56.9
52.1
51.1
55.8
51.2
53.2
45.1
59.8
55.2
53.3
65.3
19
20
21
22
23
12.40
13.00
10.00
9.70
11.90
68.6
53.2
59.9
59.3
62.4
10 pg/mL
10,47 pg/mL
1,55
14.83
60 pg/mL
56.36 pg/mL
5.32
9.43
Target cone.
Mean
SD
CV (%)
646
THCA production from smoke or oral fluid bioconversion
Production of THCA as a pyrolysis product of cannabis
smoke or as a result of bioconversion in the oral fluid was
briefly investigated.Metabolismof THC in oral fluidwas briefly
investigated to rule out conversion of TIqCto THCAin the oral
fluid. This was done by incubating THC at 1.0 mg/mL in oral
fluid at 37~ for 24 h and then testing for the formation of
THCA. Production of THCA from cannabis smoke was also
briefly investigated. A volunteer participating in a cannabis
smoking study inhaled smoke from a lit marijuana cigarette
and exhaledthis smoke into an aqueous solution. This solution
was tested for THC and THCAby GC-MS-MS.
Accuracy
Accuracyof the analytical method was determined at 10, 20,
40, 60, and 120 pg/mL by extracting fivesamples at each concentration. The average response was calculatedas a percent of
the expected response and reported with the relative standard
deviation. Precision was calculated as the relative standard
deviation of the sample mean of these replicates across this
range. Between-run precision was calculated at 10 and 60
pg/mL by collecting data from 23 batch runs performed on separate days.The 10 and 60 pg/mL control from each batch of 23
were averagedand the relative deviationwas calculated and reported as the between-run precision at each concentration.
Linearity
Linearity of sample response was evaluatedin a range below
250 pg/mL. This was performed by spiking known amounts of
THCAin oral fluid and subjecting these samples to the analysis
described. A correlation analysis was then performed, and the
r 2 for detector response versus analyte concentration was reported. Repeated batches were plotted together in the low
range of 10 to 60 pg/mL.
Recovery
Analyticalrecovery or extraction efficiencywas performedat
10, 20, and 60 pg/mL. Analytical recovery experiments were
performed by first extracting the THCAdrug then adding internal standard after the extraction.
THCA-glucuronideliberation
Additional studies were performed to determine if THCAglucuronide was hydrolyzedto liberate freeTHCAusing the alkaline extraction procedure. THCA-glucuronidewas purchased
from Alltech Applied Science (State College, PA) and was determined to be greater than 99% pure when assayedfor THCA
by the GC-MS-MS described here. A sampleof the glucuronide
was dried and derivatizedwithout the hydrolysisstep in the following protocol to determine the free THCAcontent. THCAglucuronide was also subjected to hydrolysisand extraction to
verify recovery of free THCA.
Sample preparation
Solid-phase extractions were performed without column
conditioning. One milliliter of 1.0N NaOH was added to
CEREX| Polychrome extraction columns (CERA, Baldwin
Park, CA).Then 0.1 mL of the oral fluid, calibrator, or control
Journal of Analytical Toxicology, Vol. 30, November/December2006
was added, followedby THCA-d3internal standard. These mixtures were then passed through the extraction columns at a
low flow rate with positive pressure using nitrogen gas. Before
allowing the column to dry, the column was washed with 1.0
mL of a mixture of water/methanol/ammonium hydroxide
(85:15:1) followed by 0.5 mL methanol. The columns were
dried for 15-30 rain with nitrogen gas using positive pressure.
Elution of column bound contents was performed using a
solvent mixture of hexane/ethyl acetate/acetic acid (80:20:2).
Following this elution, the eluate was dried completely under
a stream of nitrogen at 40~ Fifty microliters of HFIP and
PFAAwere added, and the vial was capped and heated at 75~
for 15 rain. Derivative was dried again using nitrogen, and 50
IJL toluene was added to the residue. After brief mixing, the
toluene was then transferred to a 0.7-mL amber glass autosample vial for injection.
Sampleanalysis
Sample analysis was performed on a Finnigan TSQ-7000
triple-quadrupole MS coupled with a ThermoFinnigan Trace
GC and a Leap Technologies CTC A200S autosampler. The instrument operated in negative ion mode, using ammonia gas
at a pressure of approximately 7 Torr. Collision gas was argon
with a pressure of 1.0 mTorr. Source temperature was held
constant at 180~ Injection volume was 3 IJL. The temperature of the injector was 250~ and the GC oven temperature
program started at 130~ and ramped to 300~ using a gradient of 30~
The column used was a J&W DB-5 (15 m x
0.25-ram i.d., 1.0-1Jm film thickness, Agilent Technologies,
Palo Alto, CA).The transfer line temperature was set at 270~
Tuning of the instrument was performed prior to each set of
analysis.
Retention times for each sample peak were within _+2% of
the retention time for the calibrator. Precursor ions monitored for THCA and THCA-d3were m/z 620.5 and 623.5, respectively. Product ions were m/z 492 and 383 for THCAand
495 and 386 for THCA-d3.Quantitation was performed using
the internal standard peak-area ratios of the m/z 495 peak
with the m/z 492 peak. Calculations were performed using
Finnigan Excaliber software and Microsoft Excel for ion ratio
calculations. Acceptance of unknown samples required retention time and ion ratio match. Ion ratio acceptance of each unknown sample required the presence of a peak at the correct
retention time for the 492 product ion and the 383 product
ion; moreover, the ion ratio of 492/383 was within 30% of the
ion ratio of the calibrator. Samples that contained both 492 and
383 peaks with greater than 3:1 signal to noise, but did not
meet the 30% ion-ratio criteria were below the LOQ.
phine, methadone, lidocaine, naproxen, chlorpheniramine,
erythromycin, diazepam, pentazocine, imipramine, phenolphthalein, doxepin, carbamazepine, quinine, methaqualone,
diphenhydramine, cocaine, amoxapine, desipramine, benzoylecgonine, propoxyphene, trazodone, and phentermine.
Concentration ranges were from 100 ng/mL to 10 I~g/mL.
Stability
Short-term sample stability was briefly investigated, and additional experiments are currently underway.Authentic samples
were placed on stability refrigerated stored in amber glass vials.
THCA-spikedoral fluid samples stored in polypropylene tubes
(Dependable Scientific, Midvale, UT) were evaluated at two
temperature conditions: room temperature and refrigerated.
THCA-glucuronide spiked oral fluid samples stored in
polypropylene tubes were evaluated at two temperature conditions: room temperature and frozen. Derivatized analyte was
studied for stability over a 72-h period stored at room temperature in amber glass vials to determine stability of the analyte
after extraction and derivitization, and before analysis.
Results and Discussion
To our knowledge, this is the first published report of a
method for quantitative determination for THCAin oral fluids
from marijuana users. This method consists of alkaline hydrolysis followed by solid-phase extraction and GC-MS-MS
analysis for the detection and quantitation of THCA.The assay
is linear from 10 to 240 pg/mLwith a sample volume of 100 IJL
oral fluid. This represents a low of 60 fg of derivatized THCAon
column with a signal-to-noise ratio above three for the qualifier ions.
Oral fluid samples
THCA was detected in 81% of the oral fluid samples in
Group C (screened and confirmed positive for THC) with the
remaining samples containing THCA below the cutoff of 10
pg/mL. All samples in Group C were from long-term storage of
donor samples that had positive THC results previously reported. Samples ranged from the lower limit of detection of 10
pg/mL to a high of 142 pg/mL. THCAwas detected in 1 sample
from Group A (screened negative) at 15.7 pg/mL, and 2 samples in Group B (screened positive) at 32.2 and 82.0 pg/mL. The
data show that the current oral fluid cutoff used for THC correlates with THCA detection at 10 pg/mL. Table II shows the
THC concentrations determined in the oral fluids and the corresponding THCA determined for those screened and confirmed positive for THC.
Specificity
Specificity experiments were performed to verify that no
interference was detected by subjecting a known library of
compounds to this assay. Compounds used for specificity experiments included THC, 11-hydroxy-THC, acetaminophen,
meperidine, d-methamphetamine, carisoprodol, meprobamate,
PPA, hydroxyzine, nortriptyline, PCP, caffeine, amitriptyline,
nicotine, pseudoephedrine, d-amphetamine, codeine, mor-
THCA production from THC
Blown cannabis smoke did not contain detectable THCA.No
formation of THCAwas detected when incubation of THC in
oral fluid was performed. Further, it is unlikely that any biotransformation in the oral mucosa could be responsible for
converting THC to THCA. Several authors have reported that
the enzymes responsible for this conversion, are hepatic re-
647
Journal of Analytical Toxicology, Vol. 30, November/December 2006
lated and require several microsomal enzyme pathways (8-12).
Microsomal hydroxylation allylic to A9-THC double bond occurs to form ll-OH-A9-THC. The conversion to THCA is accomplished by alcohol dehydrogenase (12), which is then
subsequently glucuronidated (11). None of these enzymes are
known to exist in the oral mucosal. The presence of THCA in
oral fluid may come from diffusion of THCAfrom the blood or
transport by some other mechanism. The low levels of THCA
detected in oral fluid are consistent with its high degree of
plasma protein binding, and are below the levels of detection
for validated methods previously reported in the literature.
240 pg/mL. Chromatograms are shown in Figure 1. The linearity curve is shown in Figure 2.
Recovery
Analytical recovery of freshly spiked oral fluid is tabulated
in Table III. At the LOQ of 10 pg/mL recovery was 113%
m/z 383
80-
1805
80
Accuracyaverage in the range of 10 to 120 pg/mL was 108%
with a relative standard deviation of 15%. Precision of repeated
extractions on 23 different batch runs of the same concentration was 16.6% at the LOD (10 pg/mL) and 9.38% at 60 pg/mL.
Precision within a batch run (n = 5) was 20% at 10 pg/mL, 4%
at 20 pg/mL, and 7% at 60 pg/mL. Analytical recovery/extraction efficiency was 113% at 10 pg/mL, 92.5% at 20 pg/mL,
and 105% at 60 pg/mL. Between-run accuracy and precision at
10 pg/mL and 60 pg/mL are shown in Table I.
I
.Q,
G0
6o-"
5/{/
4(3
20
20
"~--~-~
rn
~J
Time(rain
k
Time (ruin)
RT: 5.54
S#: t 728
AA: 46837828
AH: 27905014
1oo-4
i 834
1810 5.75
1O0
5,N h
mlz 495
I
1778 1
i5o
4O
Fluid Samples
Analyzed for THC
1726 5.83j
20
"i i i , "j ~,7 ~ ' ~ , ,
5,4
5.5
0
5.6
'
THC (ng/mL)
THCA (pg/ml)
331
678
203
422
686
841
071
451
452
085
410
214
413
808
474
080
982
528
989
300
020
591
258
677
521
048
0.5
0.7
0.9
1.0
1.0
1.0
1.2
1.5
1.7
2.9
6.6
9.8
10.6
10.7
11.7
11.9
14
17
18.6
22.5
24
25
28
29.4
31.9
57
52
15
11
11
11
< 10
16
52
17
35
< 10
27
< 10
15
142
37
11
69
61
71
< 10
< 10
10.8
32
78
32
* Lower limits of quantitation were 0.4 ng/mL for THC and 10 pg/mL for THCA.
' I ' ' ' ' I
5.5
Time (role)
LabID
\
iI '
20
Table II. Authentic Oral
and THCA*
f~/\
mlz 492
80'--[
Linearity
The correlation coefficientfor assay linearity exceeded0.995
between the LOQ of 10 pg/mL and the high concentration of
,,,,
40
801
648
101
mlz 386
Accuracy
A
Negative control
RT: 5.53
S#: 1727
Ag: 77121141
AH: 44540253
'''
' I ' ' '
Low control
10.0 pg/mL TttCA in oral fluid
Rr: 5.53
S#: 1727
AA: 137338720
AH: 77323182
100m
8o
40
'
5,8
B
RT: 5.54
S#: 1729
MA: 1332902
MH: 719832
100 -,
88
'l
5.6
5.7
Time (ml~)
m/z383
1661
20
~' 20-3-
,~
o-:[
5.4
Time ( ~ n )
RT: 5.53
s 1728
AA: 95308717
AH; 55586246
//1/Z 492
8 80-"
60
i
40
~ 40-
2
~ 20i,
,l
11
ii,
5.5
Time (rain)
,i
5.8
, , ,
5,54
S#: 1730
MA: 986573
MH: 632908
100-
80
5,4
5.5
S ,6
Time (n~In)
RT:
~z4
,~
Y
1778
60:
1670
/4~,~//'~"
.ZJ
0-
'514....
5s .... 61o"
Time (mln)
Figure1. Chromatogramof negative oral fluid with internal standard;
parent mass is m/z 623 with fragments m/z 386 and 495 (A). Chromatogram of control at 10 pg/mL THCA; parent mass is m/z 620 with
fragments m/z 383 and 492 (B).
Journal of Analytical Toxicology, Vol. 30, November/December2006
with a relative standard deviation of 16%. At 20 pg/mL,
the concentration of the calibrator, the average recovery
was 100% with a relative standard deviation of 2%. And
at 60 pg/mL, the average recovery was 120% with a relative
standard deviation of 15%. In this low range, the method
meets current requirements listed in the Center for Drug
Evaluation and Research (CDER) for recovery of analyte of
interest (13).
THCA-glucuronide liberation
THCArecovery from the THCA-glucuronidespiked oral fluid
was good. Three samples of THCA-glucuronide were spiked at
66 pg/mL THCA content. Recoveries of these three samples
were 64.47, 70.80, and 52.65 pg/mL, or 94.9% on average. The
THCA-glucuronide, if present in the oral fluid, was measured
and reported as free THCA. Because of the limited supply of purified THCA-glucuronide, these recovery studies were only
performed in triplicate. A larger sample size is desired before
making any conclusion about recovery statistics of the glucuronide form of THCAusing this method. THCA-glucuronide
is known to be present in the blood and urine after cannabis
use (14), but it is not known at this time if THCA-glucuronide
can be found in oral fluid.
300 -
.y
y = 1.0107x
200 -
Interference was not seen from specificity samples tested
including a library of known compounds investigated, listed
here. All negative samples of oral fluid collected from drug
free volunteers showed no interference while subjected to this
analysis.
Stability
Room temperature stability of THCA spiked in oral fluid is
known to be poor when samples are stored at room temperature (15). Great care was taken to perform method validation
with freshly prepared samples. Stability studies performed with
spiked oral fluid confirmed that THCA is not stable in oral
fluid at room temperature (Figure 3). Only 10% of the amount
spiked could be recovered after four days storage. In contrast,
stability of THCA-glucuronide does not show this same trend
(Figure 4). For a period of 60 days THCA-glucuronide samples
Room temperature stability of THCA in oral fluid
stored in polypropylene tubes (n = 4)
120
.~
100
I * pg/mL
" < ~ 80
r~.~
~ = 6O
Linearity
250 -
Specificity
-
,o
0.6726 p
I
20
1
I
I
i
150 -
0
5
10
Time (days)
~o
I00 -
I
15
1.)
Figure 3. Stability of THCA spiked into oral fluid when stored in
polypropylene tubes at room temperature (20-24~
20
50 -
1
0
0
50
100
150
200
Expected(pg/mL)
250
300
Figure2.Linearity.
120
Table III. Recovery of THCA from Freshly Spiked Oral
Fluid with Average, Standard Deviation (SD), and
Percent Relative Standard Deviation (%RSD)
Sample
10 pg/mt
20 pg/mL
60 pg/mL
nl
n2
n3
n4
n5
n6
10.6
12.2
8.2
13.4
10.8
12.6
19.4
17.2
23.2
21
19.4
20.4
60.6
66.4
78.8
92.8
71.4
67.2
Average
11.3
1.86
16.44
20.1
1.99
9.93
72.87
11.48
15.76
SD
%RSD
THCA-glucuronide stability in oral fluid stored
in polypropylene tubes (n = 3)
~,~60
4O
9Freezer
20
A 40~
0
I
0
I
20
40
Time (days)
I
60
Figure 4. Stability of THCA-glucuronide(reportedas liberatedTHCA) in
oral fluid when stored in polypropylene tubes and either frozen (-20~
or kept at 40~ for 60 days.
649
Journal of Analytical Toxicology,VoL 30, November/December2006
were stable. Authentic Samples stored in amber glass vials
were also stable for a period of five months (data not shown).
Authentic samples showed similar stability to the THCA-glucuronide samples, yet, because of limited supply of authentic
oral fluid samples and limited supply of THCA-glucuronide,
storage conditions were not identical and future stability
studies are limited by this supply. However, data clearly show
THCA in oral fluid of authentic samples well above the LOQ.
Table II shows the level of THCAdetermined in authentic samples. Some of these authentic samples had been in storage for
up to two years before this test was performed. Full conclusions
about the stability of THCAand THCA-glucuronide in oral require further investigation.
2.
3.
4.
5.
6.
7.
8.
Conclusions
THCAwas detected in oral fluid of samples previously identified as positive for THC. A method is now available to investigate the appearance of the major metabolite of cannabis in
oral fluid at low picogram-per-mil]iliter levels. The low levels
detected in authentic samples indicate previous reports may
have missed detection of THCAbecause of sensitivity.
Acknowledgments
9.
10.
11.
12.
We would like to thank Rodger L. Foltz for expert analytical
advice, consultations and references.
13.
14.
References
15.
1. R.A. Gustafson, I. Kim, P.R. Stout, K.L. Klette, M.P. George,
E.T. Moolchan, B. Levine, and M.A. Huestis. Urinary pharmacokinetics of 11-nor-9-carboxy-Ag-tetrahydrocannabinol after con-
650
trolled oral Ag-tetrahydrocannabinol administration. J. Anal. Toxicol. 28:160-167 (2004).
P.R.Stout, C.K. Horn, and K.L. Klette. Solid-phase extraction and
GC-MS analysis of THC-COOH method optimized for highthroughput forensic drug-testing laboratory. J. Anal. Toxicol. 25:
550-554 (2001).
P. Kintz. Drug Testing in Hair. CRC Press, Boca Raton, FL, 1996.
K.M. Hold, D. de Boer, J. Zuidema, and R.A.A. Maes. Saliva as an
analytical tool in toxicology. Utrecht Institute of Pharmaceutical
Sciences (UIPS), Utrecht University, Utrecht, the Netherlands.
N. Samyn, A. Verstraete, C. van Haeren, and P. Kintz. Analysis of
drugs of abuse in saliva. Forensic Sci. Rev. 11(1): 1-19 (1999).
W. Schramm, R.H. Smith, P.A. Craig, and D.A. Kidwell. Drugs of
abuse in saliva: a review. J. Anal. Toxicol. 16:1-9 (1992).
R.E.Choo and M.A. Huestis. Oral fluid as a diagnostic tool. Clin.
Chem. Lab. Med. 42(11): 1273-1287 (2004).
D.J. Harvey and N.K. Brown. Comparative in vitro metabolism of
the cannabinoids. Pharmacol. Biochem. Behav. 40(3): 533-540
(1991).
M.E. Wall and M. Perez-Reyes.The metabolism of delta 9-tetrahydrocannabinol and related cannabinoids in man. J. Clin. Pharmacol. 21 (8-9 suppl): 178S-189S (1981 ).
K. Watanabe, S. Narimatsu, T. Matsunaga, I. Yamamoto, and
H. Yoshimura. A cytochrome P450 isozyme having aldehyde
oxygenase activity plays a major role in metabolizing cannabiholds by mouse hepatic microsomes. Biochem PharmacoL 46(3):
405-411 (1993).
D.J. Harvey, B.R. Martin, and W.D. Paton. Identification of the glucuronides as in vivo liver conjugates of seven cannabinoids and
some of their hydroxyl and acid metabolites. Res. Commun.
Chem. Pathol. Pharmacol. 16(2): 265-279 (1977).
M.E. Wall, B.M. Sadler, D. Brine, H. Taylor, and M. Perez-Reyes.
Metabolism, disposition, and kinetics of delta-9-tetrahydrocannabinol in men and women. Clin. Pharrnacol. Ther. Sep.
34(3): 352-363 (1983).
FDA guidance page. http://www.fda.gov/cder/guidance/index.htm.
W. Weinmann, S. Vogt, R. Goerke, C. Muller, and A. Bromberger.
Simultaneous determination of THC-COOH and THC-COOHglucuronide in urine samples by LC/MS/MS. Forensic Sci. Int.
113(1 ): 381-389 (2000).
D.J. Crouch. Oral fluid collection: the neglected variable in oral
fluid testing. Forensic Sci. Int. 150(2-3): 165-173 (2005).
Manuscript received April 1,2005;
revision received November 1,2005.