Y. Ruben Luo (Presenter)
University of California, San Francisco
Bio: Y. Ruben Luo received his PhD in analytical chemistry from Stanford University, and his BS in chemistry from Peking University, China. He is currently a clinical chemistry fellow at the University of California, San Francisco. Prior to his fellowship, he worked in the clinical diagnostics industry for 9 years, holding positions as product manager of a global biotech / clinical diagnostics company and technical director of a third-party laboratory. His research experience focused on the development of clinical diagnostic assays and devices based on mass spectrometry and optical biosensors.
Authorship: Y. Ruben Luo, Alan H. B. Wu, Kara L. Lynch
Department of Laboratory Medicine, University of California, San Francisco and Zuckerberg San Francisco General Hospital, San Francisco, CA, USA
| Short Abstract Due to the legalization of recreational cannabis use in some states in the US, simple and rapid cannabis use monitoring under certain circumstances is in urgent demand for both public health and safety concerns. A prototype breathalyzer for real-time cannabis use monitoring was built allowing for the collection of liquid-form breath samples. Two LC-MS/MS methods with simple and rapid sample preparation procedures were developed for quantitation of 5 cannabinoids THC, THCCOOH, 11-OH-THC, CBN, CBD in both blood (serum) and breath samples. The serum samples were “crashed” to remove proteins, filtered, and directly analyzed in LC-MS/MS without sample dry-down and reconstitution. The breath samples were simply diluted and directly analyzed in LC-MS/MS. Validation of the LC-MS/MS methods was carried out in terms of precision, accuracy, linear range, limit of quantitation (LOQ), and carryover. |
Long Abstract
Introduction
Cannabis is a commonly abused drug. In the US, in addition to medical use, its recreational use has recently been legalized in several states including California. Thus, simple and rapid cannabis use monitoring under various circumstances, such as vehicle driving, is in urgent demand for both public health and safety concerns. A new type of breathalyzer has been prototyped as a tool for real-time cannabis use monitoring. It dissolves chemical components contained in 5 L exhaled breath into ethanol to form a breath sample, which can be analyzed in a point-of-care device or by mass spectrometry.
Cannabinoids in whole blood and urine samples have been extensively studied; however there are only a few reports of quantitation in breath samples. It is valuable to determine if the levels of THC in breath correlate with THC in blood samples (whole blood / serum). This exploratory research can add in the development of the new type of breathalyzer for cannabis administration monitoring.
Two LC-MS/MS methods with simple and rapid sample preparation procedures were developed for quantitation of 5 cannabinoids in blood (serum) and breath samples. The serum samples were “crashed” to remove proteins, filtered, and directly analyzed in LC-MS/MS without sample dry-down and reconstitution. The breath samples were simply diluted and directly analyzed in LC-MS/MS. Validation of the LC-MS/MS methods was carried out in terms of precision, accuracy, linear range, limit of quantitation (LOQ), and carryover.
Methods
(1) Method for Blood (Serum) Samples
Cannabinoid standards were purchased from Cerilliant: Δ9-tetrahydrocannabinol (THC, Cat# T-005), 11-hydroxy-Δ9-tetrahydrocannabinol (11-OH-THC, Cat# H-027), cannabinol (CBN, Cat# C-046), cannabidiol (CBD, Cat# C-045), all as 1.0 mg/ml solution in methanol, 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THCCOOH, Cat# T-006), as 100 μg/ml solution in methanol. Stable isotope-labeled internal standards (IS) were purchased from Cerilliant: THC-D3 (Cat# T-003), 11-OH-THC-D3 (Cat# H-041), THCCOOH-D3 (Cat# T-004), CBN-D3 (Cat# C-115), CBD-D3 (Cat# C-084), all as 100 μg/ml solution in methanol. Drug-free serum as sample matrix for calibrators and quality controls was purchased from Bio-Rad (Cat# 456).
A standard stock solution containing 10 μg/ml THC, THCCOOH, 11-OH-THC, CBN, CBD was prepared in methanol. The stock solution was diluted in drug-free serum to prepare a 200 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD calibrator (Cal 1). The Cal 1 was diluted in drug-free serum in 2-fold series to prepare Cal 2 ~ Cal 13 at concentrations of 100 ng/ml, 50 ng/ml, 25 ng/ml, 12.5 ng/ml, 6.25 ng/ml, 3.125 ng/ml, 1.562 ng/ml, 0.781 ng/ml, 0.391 ng/ml, 0.195 ng/ml, 0.098 ng/ml, 0.049 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD.
Quality controls were prepared in drug-free serum separately from the calibrators. Three levels of quality controls were made at concentrations: L1 1.0 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD; L2 5.0 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD; L3 100 ng/ml THC, THCCOOH, 50 ng/ml 11-OH-THC, CBN, CBD.
An IS stock solution containing 1 μg/ml THC-D3, THCCOOH-D3, 11-OH-THC-D3, CBN-D3, CBD-D3 was prepared in methanol. A precipitant-IS mixture was made by diluting the IS stock solution to 2.5 ng/ml in 1:1 MeOH : ACN.
In sample preparation, 50 µl of each serum sample was loaded in a 96-well filter plate (Sigma-Aldrich Cat# 55263-U), and 100 µl precipitant-IS mixture was added to precipitate proteins, followed by shaking the samples for 0.5 min on a plate shaker. Then the filter plate was put on a positive-pressure manifold to filter the samples into a deep-well plate. The deep-well plate was sealed with a pre-pieced seal to minimize sample evaporation, and loaded in a LC-MS/MS system for LC-MS/MS analysis.
LC-MS/MS analysis was carried out on a 20A HPLC (Shimadzu) and QTRAP 4500 triple-quadrupole mass spectrometer (Sciex). Mobile phase A (MPA) was 10 mM ammonium acetate in water and mobile phase B (MPB) was 1:1 MeOH : ACN. A Kinetex C18 column (3.0 mm x 50 mm, 2.6 µm particle) (Phenomenex, Cat# 00B-4462-Y0) was used with a security guard ULTRA column and holder (Phenomenex, Cat# AJ0-8775 and Cat# AJ0-9000). Gradient elution was employed in HPLC separation: equilibration at 50% B for 3 min, 50% B for 0.1 min, increased to 60% B over 0.4 min, then increased to 100% B over 4 min and held for 3 min, returned to 50% B over 0.1 min and held for 1.4 min. Sample injection volume was 15 µl, flow rate was 0.50 ml/min and column temperature was 25°C. MS conditions: ESI source, negative mode, curtain gas at 45 psi, collision gas at 10 psi, ion source gas 1 at 70 psi, ion source gas 2 at 60 psi, ionspray voltage at -4000 V, temperature at 550°C. Scheduled multiple-reaction monitoring (MRM) was set for the following transitions with cycle time 0.3 s and window width 30 s centered at retention time.
THC: m/z 313.22 → 191.11, m/z 313.22 → 245.15, retention time 4.75 min
THCCOOH: m/z 343.19 → 299.17, m/z 343.19 → 245.15, retention time 2.90 min
11-OH-THC: m/z 329.21 → 268.15, m/z 329.21 → 173.10, retention time 3.75 min
CBN: m/z 309.19 → 279.14, m/z 309.19 → 222.10, retention time 4.55 min
CBD: m/z 313.22 → 179.11, m/z 313.22 → 245.15, retention time 4.05 min
THC-D3: m/z 316.24 → 194.13, retention time 4.75 min
THCCOOH-D3: m/z 346.21 → 302.18, retention time 2.90 min
11-OH-THC-D3: m/z 332.23 → 271.17, retention time 3.75 min
CBN-D3: m/z 312.20 → 282.16, retention time 4.55 min
CBD-D3: m/z 316.24 → 182.13, retention time 4.05 min
(2) Method for Breath Samples
Cannabinoid standards and stable isotope-labeled internal standards (IS) were the same as described in Method for Blood (Serum) Samples. Ethanol as sample matrix for calibrators and quality controls was purchased from Sigma-Aldrich (Cat# 459828).
A standard stock solution containing 10 μg/ml THC, THCCOOH, 11-OH-THC, CBN, CBD was prepared in methanol. The stock solution was diluted in ethanol to prepare a 200 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD calibrator (Cal 1). The Cal 1 was diluted in drug-free serum in 2-fold series to prepare Cal 2 ~ Cal 15 at concentrations of 100 ng/ml, 50 ng/ml, 25 ng/ml, 12.5 ng/ml, 6.25 ng/ml, 3.125 ng/ml, 1.562 ng/ml, 0.781 ng/ml, 0.391 ng/ml, 0.195 ng/ml, 0.098 ng/ml, 0.049 ng/ml, 0.024 ng/ml, 0.012 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD.
Quality controls were prepared in ethanol separately from the calibrators. Three levels of quality controls were made at concentrations: L1 1.0 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD; L2 5.0 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD; L3 75 ng/ml THC, THCCOOH, 11-OH-THC, CBN, CBD.
An IS stock solution containing 1 μg/ml THC-D3, THCCOOH-D3, 11-OH-THC-D3, CBN-D3, CBD-D3 was prepared in methanol. A diluted IS stock solution was prepared by diluting the IS stock solution to 25 ng/ml in 1:1 MeOH : ACN. A diluent-IS mixture was made by mixing 1 unit of the diluted IS stock solution and 6 units of 10 mM ammonium acetate in water.
In sample preparation, 50 µl of each serum sample was loaded in a sample vial with an insert, and 70 µl diluent was added, followed by agitating the samples on a vortex mixer. The sample vials were loaded in a LC-MS/MS system for LC-MS/MS analysis.
LC-MS/MS analysis conditions were the same as described in Method for Blood (Serum) Samples except that sample injection volume was changed to 50 µl.
Results
(1) Results for Blood (Serum) Samples
Precision and Accuracy (Trueness)
Precision was determined by running 6 ~ 8 replicates of QC L1 ~ L3 samples in repeated in 6 sample batches (prepared separately). Accuracy was determined by the trueness of average quantitation results of QC samples. Outliers were excluded according to Reed’s criterion. For all analytes, accuracy ranged 95% ~ 105% at QC L1, 96% ~ 108% at QC L2, 85% ~ 95% at QC L3; within-run precision (CV) ranged 2.7% ~ 14.0% at QC L1, 0.6% ~ 10.5% at QC L2, 0.9% ~ 9.3% at QC L3; and between-run precision (CV) ranged 5.3% ~ 12.9% at QC L1, 4.1% ~ 8.6% at QC L2, 4.8% ~ 7.7% at QC L3.
Linearity was evaluated with least squares regression lines with ≥ 10 non-zero calibrators in 12 batches. LOQ was defined as the concentration with signal/noise ratio ≥ 10, symmetrical peak shape, retention time ±0.1 min mean calibrator retention time, CV ≤ 20% in a group of separately prepared samples (n ≥ 4), and result within ±20% of target. The linear range is from LOQ to 200 ng/ml for all analytes. The LOQ for each analyte is: THC 0.195 ng/ml (CV 18.6%, Accuracy 106%); THCCOOH 0.391 ng/ml (CV 9.1%, Accuracy 106%); 11-OH-THC 0.049 ng/ml (CV 16.6%, Accuracy 89%); CBN 0.098 ng/ml (CV 10.1%, Accuracy 117%); CBD 0.781 ng/ml (CV 13.2%, Accuracy 121%).
Carryover was determined by measuring the analytes carried over from the highest calibrator to an immediately following blank drug-free serum sample. The limit is set at 25% of LOQ value and all analytes met the criterion.
(2) Results for Breath Samples
Precision and Accuracy (Trueness)
Precision was determined by running 6 replicates of QC L1 ~ L3 samples in repeated in 4 sample batches (prepared separately). Accuracy was determined by the trueness of average quantitation results of QC samples. For all analytes, accuracy ranged 94% ~ 105% at QC L1, 95% ~ 104% at QC L2, 96% ~ 109% at QC L3; within-run precision (CV) ranged 2.1% ~ 12.7% at QC L1, 2.7% ~ 12.8% at QC L2, 2.4% ~ 10.6% at QC L3; and between-run precision (CV) ranged 3.4% ~ 9.7% at QC L1, 7.0% ~ 9.8% at QC L2, 4.7% ~ 9.9% at QC L3.
Linearity was evaluated with least squares regression lines with ≥ 10 non-zero calibrators in 12 batches. LOQ was defined as the concentration with signal/noise ratio ≥ 10, symmetrical peak shape, retention time ±0.1 min mean calibrator retention time, CV ≤ 20% in a group of separately prepared samples (n = 5), and result within ±20% of target. The linear range is from LOQ to 100 ng/ml for all analytes except for CBN to 50 ng/ml. The LOQ for each analyte is: THC 0.012 ng/ml (CV 9.4%, Accuracy 115%); THCCOOH 0.012 ng/ml (CV 12.2%, Accuracy 103%); 11-OH-THC 0.012 ng/ml (CV 21.1%, Accuracy 108%); CBN 0.012 ng/ml (CV 8.6%, Accuracy 87%); CBD 0.012 ng/ml (CV 7.9%, Accuracy 115%).
Carryover was determined by measuring the analytes carried over from the highest calibrator to an immediately following blank drug-free serum sample. The limit is set at 25% of LOQ value and all analytes met the criterion.
(3) Correlation between Blood (Serum) and Breath Samples
Blood and breath samples taken from 9 test subjects were analyzed using the LC-MS/MS methods for the correlation study. For each subject, samples were collected right before cannabis administration (0 min), and 15 min, 30 min, 45 min, 60 min, 90 min, 120 min after cannabis administration. THC was the main cannabinoid detected in breath samples. Positive correlation between its concentrations in serum and breath samples was observed. THC peaked at 15 min in both sample forms, and in serum samples it decreased over time to around 20% of its peak value in 2 hours while in breath samples it mostly became not detectable at 60 min. We picked 3 test subjects with relatively high THC concentrations in serum samples, and plotted the THC concentrations in breath samples at 15 min, 30 min, 45 min against those in serum samples. The correlation coefficient (r) is 0.94 and slope is 0.055 in linear regression fitting, indicating the existence of quantitative relation between breath and serum samples in general.
Conclusions & Discussion
The preliminary results showed positive correlation between the breath samples and the serum samples obtained at the same time from the same test subjects. This exploratory research is a small step in our understanding of THC in breath and has provided preliminary data that will add in the continued development of the new breathalyzer for cannabis administration monitoring.
References & Acknowledgements:
| Description | Y/N | Source |
| Grants | no | |
| Salary | yes | 10x Genomics |
| Board Member | no | |
| Stock | no | |
| Expenses | no |
IP Royalty: no
| Planning to mention or discuss specific products or technology of the company(ies) listed above: | no |