Analytical and Bioanalytical Chemistry

, Volume 405, Issue 4, pp 1399–1406

Simultaneous and sensitive LC–MS/MS determination of tetrahydrocannabinol and metabolites in human plasma

Authors

    • Pharmazentrum Frankfurt/ZAFES, Institute of Clinical PharmacologyGoethe-University Frankfurt
  • S. Labocha
    • Pharmazentrum Frankfurt/ZAFES, Institute of Clinical PharmacologyGoethe-University Frankfurt
  • C. Walter
    • Pharmazentrum Frankfurt/ZAFES, Institute of Clinical PharmacologyGoethe-University Frankfurt
  • J. Lötsch
    • Pharmazentrum Frankfurt/ZAFES, Institute of Clinical PharmacologyGoethe-University Frankfurt
  • G. Geisslinger
    • Pharmazentrum Frankfurt/ZAFES, Institute of Clinical PharmacologyGoethe-University Frankfurt
Original Paper

DOI: 10.1007/s00216-012-6501-x

Cite this article as:
Ferreirós, N., Labocha, S., Walter, C. et al. Anal Bioanal Chem (2013) 405: 1399. doi:10.1007/s00216-012-6501-x
  • 613 Views

Abstract

Cannabis is not only a widely used illicit drug but also a substance which can be used in pharmacological therapy because of its analgesic, antiemetic, and antispasmodic properties. A very rapid and sensitive method for determination of ∆9-tetrahydrocannabinol (THC), the principal active component of cannabis, and two of its phase I metabolites in plasma has been developed and validated. After solid-phase extraction of plasma (0.2 mL), the clean extracts were analyzed by tandem mass spectrometry after a 5-min liquid chromatographic separation. The linear calibration ranges were from 0.05 to 30 ng mL−1 for THC and 11-nor-∆9-carboxy-tetrahydrocannabinol (THC-COOH) and from 0.2 to 30 ng mL−1 for ∆9-(11-OH)-tetrahydrocannabinol (11-OH-THC). Imprecision and inaccuracy were always below 7 and 12 % (expressed as relative standard deviation and relative error), respectively. The method has been successfully applied to determination of the three analytes in plasma obtained from healthy volunteers after oral administration of 20 mg dronabinol.

Keywords

THCMetabolitesLC–MS/MSCannabinoids

Introduction

The European Monitoring Centre of Drugs and Drug Addiction has reported that cannabis is the most widely used illicit drug in Europe, where it is both imported and produced domestically [1]. Cannabis is commonly administered orally or by inhalation. Via its pharmacodynamic action as an agonist at cannabinoid CB1 receptors, it induces physical relaxation, changes in perception, mild euphoria, reduced motor coordination, and reduced information processing [2]. Beneficial pharmacological effects have also been demonstrated for the principal psychoactive component of cannabis, ∆9-tetrahydrocannabinol (THC) and several of its derivatives with CB1 agonist activity. These include antiemetic, appetite stimulant, and analgesic properties [3]. In this respect, oral administration of dronabinol (synthetic THC), commercialized as Marinol (Banner Pharmacaps, NC, USA), has been approved in North American and European countries for treatment of anorexia in AIDS patients, as an antiemetic during chemotherapy, and for the treatment of multiple sclerosis-associated neuropathic pain. Another derivative of THC, nabilone, has also been developed for oral administration (Cesamet; Meda Pharmaceuticals, NJ, USA) as an antiemetic. In Germany, a combination of THC and cannabidiol (CBD) is available as an oral spray for treatment of spasticity in patients suffering from multiple sclerosis [4].

THC is metabolized by cytochrome P450 isoenzymes 2C9, 2C19, and 3A4/5 [57] to several phase I metabolites. The main metabolite is the 11-hydroxy derivative of THC (11-OH-THC), which retains psychoactivity and is further oxidized to 11-nor-∆9-carboxy-tetrahydrocannabinol (THC-COOH). THC and its metabolic products also undergo phase II metabolism, which leads to the formation of glucuronide derivatives [8].

In some countries, a THC concentration above 1.0 ng mL−1 in blood is taken as constituting acute impairment of fitness to drive. Monitoring THC and THC-derivative levels also helps to establish pharmacodynamic and pharmacokinetic models to improve disease therapy with cannabis. Several methods for determination of cannabis and its metabolites in biological samples have been reported in the literature. Gas chromatographic–mass spectrometric methods for quantification and/or identification of THC and its metabolites in biological samples were published as early as the 1970s [9, 10]. The first publications dealing with determination of THC and related compounds by liquid chromatography coupled to mass spectrometry (LC–MS) date from the late 1990s [11, 12]. In recent years, application of LC–MS to the determination of THC and its metabolites has increased substantially [1317]. The principal advantages of LC–MS are that no derivatization is required and determination of glucuronide derivatives does not require expensive and time-consuming chemical or enzymatic hydrolysis (with sometimes variable efficiency of hydrolysis) [18]. Schwope et al. described a method for quantification of cannabinoids and cannabinoid glucuronides in blood samples by use of liquid chromatography–tandem mass spectrometry (LC–MS/MS) with a lower limit of quantification (LLOQ) of 1 ng mL−1 for THC, 11-OH-THC, and THC-COOH [8]. König et al. applied the same methodology and online solid-phase extraction to blood samples, achieving LLOQ of 0.5 ng mL−1 for THC and 11-OH-THC and 2.5 ng mL−1 for THC-COOH [19]. Application of this methodology to other biological samples, for example urine [2024] and saliva [25, 26] has also been reported.

The objective of this study was to develop a rapid method for quantification of THC and its phase I metabolites 11-OH-THC and THC-COOH with better sensitivity, making it particularly suitable for small volumes of plasma. The method was validated and applied to samples obtained from healthy volunteers after administration of 20 mg dronabinol [27].

Materials and methods

Materials

Acetonitrile and methanol (gradient grade) and formic acid (89–91 %) for analysis were purchased from Merck (Darmstadt, Germany). Water (LC–MS grade) and acetic acid (96 %) were obtained from Carl Roth (Karlsruhe, Germany) and AppliChem (Darmstadt, Germany), respectively.

THC, 11-OH-THC, THC-COOH and their deuterated (d3) derivatives were purchased from Lipomed (Weil am Rhein, Germany).

Plasma blank samples were obtained from healthy volunteers.

Instrumentation

Sample analysis was performed by liquid chromatography–electrospray ionization–tandem mass spectrometry (LC–ESI–MS/MS). The LC equipment consisted of an Agilent 1200 series binary pump, column oven, and degasser (Waldbronn, Germany) connected to an HTC PAL autosampler (Chromtech, Idstein, Germany). An AB Sciex (Darmstadt, Germany) 5500 QTRAP triple-quadrupole mass spectrometer equipped with a Turbo V IonSpray source operated in positive ESI mode was used for detection. High purity nitrogen for the mass spectrometer was produced by an NGM 22-LC/MS nitrogen generator (CMC Instruments, Eschborn, Germany).

Chromatographic conditions

Chromatographic separation was obtained under gradient conditions on a Mercury Luna C18 (2) column (20 × 2 mm I.D., 5 μm particle size) with a C18 guard column (4 × 2 mm I.D.) (Phenomenex, Aschaffenburg, Germany). The mobile phase was prepared from water:formic acid 100:0.1 (v/v) (component A) and acetonitrile:formic acid 100:0.1 (v/v) (component B). The gradient was: from t = 0 to 0.5 min 60 % A; from t = 0.5 to 2 min, a linear gradient from 60 to 10 % A; from t = 2 to 2.8 min, a linear gradient to 5 % A; from t = 2.8 to 3.5, 5 % A; from t = 3.5 to 4 min, a linear gradient from 5 to 60 % A, which was maintained for 1 min to re-equilibrate the system. The flow rate was 550 μL min−1. Sample extracts (10 μL) were injected for LC–ESI–MS/MS analysis.

The mass spectrometer was operated in the positive-ion mode with an electrospray potential of 5,500 V at 500 °C. Pressures of auxiliary gases 1 and 2 and the curtain and collision gases were 40, 70, 35, and 9 psi, respectively.

The precursor-to-product ion transitions m/z 315.2 → 193.0 for THC (collision energy 33 V), m/z 331.2 → 193.2 for 11-OH-THC (35 V), m/z 345.1 → 193.0 for THC-COOH (38 V), m/z 318.2 → 196.2 for d3-THC (33 V), m/z 334.2 → 316.1 for d3-11-OH-THC (30 V), and m/z 348.1 → 330.2 for d3-THC-COOH (30 V), were used as quantifiers for multiple reaction monitoring (MRM). The transitions m/z 315.2 → 123.0 for THC (47 V), m/z 331.2 → 313.1 for 11-OH-THC (30 V), and m/z 345.1 → 327.2 for THC-COOH (22 V) were used as qualifiers. For all transitions the dwell time was 50 ms.

All quadrupoles were operated at unit resolution. Quantification was performed with Analyst Software V1.5 (AB Sciex) using the internal standard method.

Standard preparation

Calibration curves from 0.05 to 30 ng mL−1 (0.05, 0.1, 0.2, 0.5, 0.75, 1.5, 3, 5, 10, 15, and 30 ng mL−1) for THC and THC-COOH and from 0.2 to 30 ng mL−1 (0.2, 0.5, 0.75, 1.5, 3, 5, 10, 15, and 30 ng mL−1) for 11-OH-THC were obtained by analysis of blank plasma spiked with 10 μL of appropriate methanolic working solutions. These working solutions were prepared by dilution of the stock solutions of the analytes (1 mg mL−1) in methanol.

Sample preparation

Plasma (200 μL) was mixed with 50 μL internal standard solution (100 ng mL−1 d3-THC and 250 ng mL−1 d3-11-OH-THC and d3-THC-COOH), 10 μL methanol (or standard solution for calibrators and quality control samples), and 400 μL 0.1 mol L−1 acetic acid. The samples were then extracted by solid-phase extraction as described elsewhere by Weinmann et al. [21]. Briefly, Chromabond C18 cartridges (500 mg, 3 mL) (Macherey–Nagel, Düren, Germany) were conditioned with 2 mL methanol and 2 mL acetic acid 0.1 mol L−1. The samples were applied to the cartridges and, after washing with 1 mL acetic acid 0.1 mol L−1 and 1 mL 70 % acetonitrile, they were dried for 7 min then eluted with 1.5 mL acetonitrile. The eluates were evaporated under nitrogen at 45 °C and reconstituted in 50 μL water:acetonitrile 60:40 (v/v) containing 0.1 % formic acid.

Method validation

The method was validated for linearity, precision, accuracy, selectivity, stability, lower limit of quantification, and limit of detection, for human plasma, in accordance with the guidelines of the Food and Drug Administration (FDA) [28, 29] and the International Conference on Harmonisation (ICH) [30].

Absolute and relative recovery of the analytes were calculated as the percentage which could be extracted from a spiked sample (A), in comparison with a neat solution of the compounds (B) and with extracted blank samples spiked after extraction (C) with the same quantity of the analytes. The internal standard (IS) was added after the extraction procedure. Recovery was assayed at three concentrations (0.05, 9, and 25 ng mL−1 for THC and THC-COOH and 0.2, 9, and 25 ng mL−1 for 11-OH-THC) and five replicates of each concentration.

Matrix effects were evaluated as suggested by Matuszewski et al. [31], by using the same set of experiments as described above. The matrix effect (ME) value (expressed in %) was calculated as: ME = C/B × 100. ME values >100 indicate enhancement whereas those <100 are indicative of ion suppression.

The calibration curves were obtained by analysis of a blank plasma sample (matrix sample processed without internal standard), a zero sample (matrix sample processed with IS), and 11 non-zero samples covering the range of 0.05–30 ng mL−1 for THC and THC-COOH and nine non-zero samples from 0.2 to 30 ng mL−1 for 11-OH-THC. Six linear calibration curves for each matrix studied were constructed by plotting the corrected areas for each concentration versus the nominal concentration of each calibration standard, applying the selected weighting factor.

The selectivity of the method was tested by comparing the signals obtained for each analyte (peak area) in blank samples (six samples) with those obtained from samples spiked at a concentration corresponding to the LLOQ.

LLOQ was calculated by interpolation of the value corresponding to 10 times the signal-to-noise ratio for drug free matrix samples in the calibration curve. The values obtained were validated by determination of relative standard deviation (RSD) and relative error (RE). The RSD, defined as (SD/mean value) × 100, where SD is the standard deviation of the measurements, was expressed as a percentage. The relative error, which is the deviation of the calculated concentration from the real spiked concentration and is calculated as ((calculated concentration − nominal concentration)/ nominal concentration) × 100, was also expressed as a percentage.

Three samples, containing low (0.05 and 0.2 ng mL−1), medium (9 ng mL−1), and high (25 ng mL−1) concentrations of THC and its metabolites were assayed in sets of five replicates to evaluate the intra-day and inter-day accuracy and precision. This procedure was repeated on three different days. The accuracy was measured as the relative error. The precision was measured as relative standard deviation.

Samples in human plasma (obtained after collection of blood in Na-EDTA tubes) were tested at two different concentrations (1 and 20 ng mL−1) for long and short term stability and at three different concentrations (0.05 (0.2 for 11-OH-THC), 9, and 25 ng mL−1) for autosampler and freeze–thaw stability. The spiked blank plasma samples were stored in 1.5-mL polypropylene tubes. The long term stability of the samples during storage at −80 °C was assayed for a period of 90 days. To assess short-term stability, samples were spiked with test compounds and stored for 5 h at room temperature.

Three replicates were analysed for each concentration.

Samples for freeze–thaw stability were spiked and immediately frozen at −80 °C. After 12 h, the samples were thawed at room temperature and refrozen at −80 °C. After the third thawing cycle, samples were extracted and measured against a freshly prepared calibration curve.

All samples were extracted after the different stability tests and cannabinoids measured against a freshly prepared calibration curve.

To demonstrate the stability of the stock solutions of all the analytes, including the internal standards, these were stored for 6 h at room temperature. The samples were then diluted to 100 ng mL−1 with methanol and measured against a freshly prepared solution.

Application of the method to real samples

Plasma was available from 30 subjects who had received an oral dose of 20 mg dronabinol in a study focussed on the pharmacodynamic effects of THC [27]. The study was performed in accordance with the Declaration of Helsinki. It was approved by the Ethics Committee of the Medical Faculty of the Goethe-University Frankfurt am Main, Germany, and all subjects had provided written informed consent. Venous blood samples (10 mL each) were collected into an Na-EDTA tube at baseline and 2 and 3 h after THC or placebo administration. Immediately after sampling, they were centrifuged at 3,000 rpm for 10 min, and the plasma was stored frozen at −80 °C pending analysis.

Results

Extraction procedure and matrix effects

Relative and absolute recovery of THC and the metabolites studied were approximately 85 % in all cases. Matrix effects ranged from 81 to 105 %. THC ionization was found to be less sensitive to matrix than for 11-OH-THC and THC-COOH, for which matrix effects were similar. In all cases, slight suppression was detected; no enhancement effects were observed. The recovery and matrix effect results obtained for the three analytes are shown in Table 1.
Table 1

Recovery (%) and matrix effects (%) (n = 5) for THC, 11-OH-THC, and THC-COOH

Compound

Concentration (ng mL−1)

Relative recovery

Absolute recovery

Matrix effects

THC

0.05

88.3 ± 3.3

86.0 ± 3.2

91.7 ± 6.7

9

82.1 ± 2.2

83.7 ± 2.3

104.8 ± 11.8

25

83.9 ± 3.4

85.4 ± 3.4

100.5 ± 3.7

11-OH-THC

0.20

89.4 ± 2.2

81.4 ± 2.04

81.1 ± 6.1

9

83.5 ± 2.1

85.6 ± 2.2

91.0 ± 6.3

25

82.8 ± 2.2

88.0 ± 2.4

88.9 ± 6.4

THC-COOH

0.05

87.4 ± 1.5

89.6 ± 1.5

87.8 ± 6.5

9

85.8 ± 2.7

83.6 ± 2.9

88.5 ± 2.6

25

85.4 ± 2.4

85.5 ± 2.4

93.0 ± 3.5

Linearity and calibration range

For each measurement, a linear calibration curve with a weighting factor of 1/x was constructed for human plasma. No significant differences were observed for any of the analytes in the different plasma samples. The calibration ranges for THC and THC-COOH were distributed along eleven calibration points and for 11-OH-THC along nine calibration points, starting with the LLOQ. The same blank plasma was spiked with all the analytes simultaneously, but for 11-OH-THC only nine of the eleven calibrators could be included in the linear regression, because of the lower sensitivity of the analytical method for this compound. The regression data obtained for six calibration curves constructed for plasma samples obtained from six different sources are presented in Table 2.
Table 2

Linearity (11 calibration points for THC and THC-COOH, 9 for 11-OH-THC)

 

Slope

Intercept

r

THC

5.1 × 10−2

6.2 × 10−4

0.9999

4.8 × 10−2

2.4 × 10−3

0.9995

5.0 × 10−2

6.2 × 10−4

0.9999

4.9 × 10−2

1.8 × 10−3

0.9995

5.1 × 10−2

1.2 × 10−3

0.9997

5.2 × 10−2

2.0 × 10−3

0.9992

11-OH-THC

1.2 × 10−2

−2.3 × 10−4

0.9997

1.1 × 10−2

2.5 × 10−4

0.9998

1.1 × 10−2

3.0 × 10−4

0.9999

1.2 × 10−2

1.1 × 10−4

0.9996

1.2 × 10−2

3.6 × 10−4

0.9999

1.1 × 10−2

−2.2 × 10−5

0.9993

THC-COOH

5.8 × 10−3

−1.1 × 10−5

0.9995

5.5 × 10−3

8.1 × 10−6

0.9991

5.6 × 10−3

8.0 × 10−5

0.9996

5.7 × 10−3

4.4 × 10−5

0.9999

5.9 × 10−3

−3.8 × 10−5

0.9996

5.6 × 10−3

3.0 × 10−4

1.0000

Selectivity

After integration of the area for each quantitative transition at the retention time corresponding to the chromatographic peaks of the analytes, it was observed that peak area for the blank samples was always less than 7.1 % of that for THC, less than 2 % of that for 11-OH-THC, and less than 10.7 % of that for THC-COOH, at the concentration corresponding to the LLOQ.

Lower limit of quantification

The concentrations set as the lower limits of quantification were chosen on the basis of signal-to-noise ratio, RE, and RSD (<20 %). For THC, the RE corresponding to 0.05 ng mL−1 ranged from 9.3 to 11.3 % and the RSD from 4.1 to 6.5 %. For THC-COOH (0.05 ng mL−1), RE values ranged from 3 to 11.9 % and RSD values, from 2.8 to 3.8 %. Finally, for 11-OH-THC (0.2 ng mL−1), RE values ranged from 8.2 to 10.5 % and RSD values, from 2.9 to 6.9 %. Chromatograms obtained for the LLOQ levels of THC and its metabolites are shown in Fig. 1.
https://static-content.springer.com/image/art%3A10.1007%2Fs00216-012-6501-x/MediaObjects/216_2012_6501_Fig1_HTML.gif
Fig. 1

Chromatograms corresponding to the LLOQ level of the three analytes (0.2 ng mL−1 11-OH-THC and 0.05 ng mL−1 THC and THC-COOH) in spiked (solid line) and blank (dotted line) plasma extracts

Precision and accuracy

The intra-day and inter-day precision (RSD) and accuracy (RE) of the method were suitable for analysis of all the samples studied. Very good precision was achieved for measurement of THC and its metabolites in plasma, with RSD values always <7 %. Inaccuracy ranged from 3 to 11.9 %.

Intraday and interday values of accuracy and precision for THC and its metabolites in plasma are collected in Table 3.
Table 3

Precision (RSD) and accuracy (RE) values for quantification of THC and its metabolites in plasma (n = 5). Values are expressed as percentages and indicate the variation and the deviation from the real concentration (100 %), respectively

 

Concentration (ng mL−1)

Day 1

Day 2

Day 3

Interday (n = 3)

RSD

RE

RSD

RE

RSD

RE

RSD

RE

THC

0.05

5.5

11.3 ± 1.6

4.1

9.3 ± 1.4

6.5

10.9 ± 2.6

5.4 ± 1.2

10.5 ± 1.1

9

2.2

5.3 ± 2.1

1.8

6.2 ± 1.9

2.2

5.2 ± 2.3

2.1 ± 0.3

5.6 ± 0.6

25

1.8

6.9 ± 1.5

3.3

5.2 ± 1.8

2.1

6.2 ± 1.8

2.4 ± 0.8

6.1 ± 0.9

11-OH-THC

0.2

6.4

10.5 ± 3.2

6.9

8.2 ± 1.7

2.9

8.6 ± 2.4

5.4 ± 2.2

9.1 ± 1.2

9

2.8

4.2 ± 2.0

1.9

6. 7 ± 0.9

2.1

5.5 ± 1.6

2.3 ± 0.5

5.5 ± 1.2

25

1.7

3.8 ± 1.3

1.8

5.8 ± 1.1

1.5

6.6 ± 0.5

1.7 ± 0.1

5.4 ± 1.4

THC-COOH

0.05

3.8

11.9 ± 0.6

3.3

10.0 ± 2.8

2.8

3.0 ± 1.5

3.3 ± 0.5

8.3 ± 4.7

9

2.8

9.7 ± 2.6

2.7

8.7 ± 3.0

1.2

9.0 ± 1.3

2.3 ± 0.9

9.1 ± 0.5

25

1.2

3.8 ± 1.3

1.0

5.1 ± 1.0

0.9

8.1 ± 0.9

1.0 ± 0.2

5.6 ± 2.2

Stability

The stability of the analytes in plasma was demonstrated at two different concentrations during 5 h at room temperature and after storage at −80 °C for 90 days. The results obtained are collected in Table 4.
Table 4

Stability of plasma samples spiked with the analytes after storage for long and short periods (n = 3). The values express the variation (%) between the real value (100 %) and the concentration measured after storage

 

Concentration (ng mL−1)

Long time (6 days)

Long time (20 days)

Long time (60 days)

Long time (90 days)

Short time (5 h)

THC

1

6.6 ± 1.5

5.1 ± 1.3

2.9 ± 1.6

13.7 ± 0.6

2.2 ± 1.2

20

4.8 ± 1.6

1.5 ± 0.5

1.5 ± 1.0

14.3 ± 0.5

2.2 ± 1.7

11-OH-THC

1

6.5 ± 1.4

10.9 ± 2.7

3.5 ± 2.2

5.9 ± 2.2

3.1 ± 1.6

20

2.4 ± 1.3

7.0 ± 0.3

4.3 ± 1.6

11.7 ± 2.1

4.1 ± 1.1

THC-COOH

1

3.0 ± 1.7

8.6 ± 1.6

4.6 ± 1.0

2.0 ± 0.5

6.8 ± 3.2

20

4.3 ± 1.5

5.0 ± 0.3

3.5 ± 1.8

7.7 ± 0.8

2.4 ± 1.2

Extracts were shown to be stable in the autosampler at 4 °C for at least 24 h and after three freeze–thaw cycles. For this purpose, three different concentrations were studied. In all cases, the variation between measured and real concentrations was <8 %.

Stock solutions, also, were stable for at least 3 months when stored at −20 °C. The stability of these solutions was tested after measured storage times by comparison with a freshly prepared solution. Variation <3 % was observed for the three compounds and their internal standards.

Application of the method to real samples

The method was used for determination of THC, 11-OH-THC, and THC-COOH in plasma from 30 healthy volunteers (fifteen women; age 27.4 ± 2.9 years, mean ± standard deviation; all within ±10 % of their ideal body weight), to whom 20 mg dronabinol had been administered orally. Samples for analysis were taken predose, and 120 and 180 min after intake of the drug. In none of the predose samples were THC or any of its studied metabolites detected. In the postdose samples, measured concentrations ranged from 0.5 to 15.9 ng mL−1 for THC, from 0.46 to 12.8 ng mL−1 for 11-OH-THC, and from 3.76 to 79.4 ng mL−1 for THC-COOH. For quantification of THC-COOH, some samples had to be diluted.

Chromatograms obtained from the plasma samples of one volunteer, before and after dronabinol intake, are shown in Fig. 2.
https://static-content.springer.com/image/art%3A10.1007%2Fs00216-012-6501-x/MediaObjects/216_2012_6501_Fig2_HTML.gif
Fig. 2

Chromatograms obtained from extracts of plasma from a healthy volunteer to whom 20 mg dronabinol was administered orally. The samples were taken predose, and 120 min and 180 min after drug intake

Conclusions

The method described was validated in accordance with international guidelines and has been successfully applied to the determination of THC and its analytes in plasma samples from healthy volunteers after oral administration of 20 mg dronabinol. Although a relatively high dose of dronabinol (20 mg) was administered to ensure pronounced pharmacodynamic effects [27], measured concentrations were within the calibration range for THC and 11-OH-THC, whereas for THC-COOH half of the samples had to be diluted before analysis. These results are in accordance with those found in the literature. Wall et al. [32] described plasma THC concentrations of 4.3 ± 0.6 and 7.1 ± 4.9 (ng mL−1 ± SD), 11-OH-THC concentrations of 7.2 ± 1.8 and 8.5 ± 2.0 (ng mL−1 ± SD) and THC-COOH concentrations of 54 ± 18 and 62 ± 17 (ng mL−1 ± SD) 120 and 180 min, respectively, after oral intake of dronabinol. Also, maximum plasma concentration of 14.5 ± 9.7 (ng mL−1 ± SD) and 9.4 ± 4.5 (ng mL−1 ± SD) were reported after oral intake of 20 and 15 mg ∆9-tetrahydrocannabinol, respectively [33].

LC–MS/MS was chosen for this study because, as described in the literature, the necessary derivatization steps at high or low pH required for GC–MS could hydrolyze several THC metabolites, for example glucuronides. This results in overestimation of the concentration of the analytes because of the contribution of the hydrolyzed compounds when comparing GC–MS and LC–MS/MS [19]. The solid-phase extraction performed before chromatographic analysis provides clean extracts and reduces interferences from sample components. No important matrix effects on the ionization of THC and its metabolites could be detected when studying the effect of human plasma samples obtained from different sources. The main effect of the matrix was on 11-OH-THC, for which slight suppression of 13 % (mean value) was observed.

The main advantage of the proposed method is its sensitivity. Use of only 0.2 mL plasma results in lower limits of quantification of 0.05 ng mL−1 (2 pg on column) for THC and THC-COOH and 0.2 ng mL−1 (8 pg on column) for 11-OH-THC. Most published methods use volumes between 0.5 to 1 mL and describe LLOQ values of 0.5 to 1 ng mL−1 for THC [8, 17, 19]. Dubois et al. developed a faster method for separation for THC and its metabolites (3 min) but this was less sensitive [34]. The sensitivity achieved by use of the methodology and instrumentation presented in this study is even higher than that achieved when using GC–MS techniques [35, 36]. In addition, to its sensitivity, the method was also very fast. Chromatographic separation took only 5 min, including re-equilibration time. Sample preparation by solid-phase extraction, currently performed manually, can, however, easily be automated for routine analysis, further reducing the time of analysis.

Application of the method described here was demonstrated for a dronabinol study of healthy volunteers; it could, because of its high sensitivity, also be used for other purposes. One possible application could be quantification of the analytes in different matrices (for example bile or vitreous humor) to improve knowledge of post-mortem redistribution of the drug and its metabolites and for identification of appropriate specimens for interpretation purposes. Another future application could be environmental analysis of THC residues.

Acknowledgments

This study was supported in part by the “Landesoffensive zur Entwicklung wissenschaftlich-ökonomischer Exzellenz: LOEWE. Schwerpunkt: Anwendungsorientierte Arzneimittelforschung” (GG, JL), the “Deutsche Forschungsgemeinschaft, DFG Lo 612/10-1 (JL)”, and the European Graduate School GRK757 (GG, JL).

Copyright information

© Springer-Verlag Berlin Heidelberg 2012