Podium Presentations for Troubleshooting

Get Your Reference Right!: Synthesis, Certification, and Confirmation of the First Reference Material for (-)-trans-11-nor-9-Carboxy-Δ9-THC beta-d-glucuronide Raymond Suhandynata (Presenter) University of California, San Diego

To be presented in Track 3 (De Anza 3) on Wednesday at 14:40

Introduction
(-)-trans-delta9-Tetrahydrocannabinol ((-)-Δ9-THC), the main psychoactive constituent of cannabis sativa L., is currently used medicinally and recreationally. The plant constituent and bioactive THC form is a specific optically active (-)-trans isomer with stereochemistry (6aR,10aR). Metabolic glucuronidation results in diastereomers with a combination of the beta anomers at the glucuronic linkage, as well as chiral centers at the 6 and 10 positions on the THC molecule. In response to the growing need to test for cannabis use in clinical and forensic settings, cannabinoid Certified Reference Materials (CRMs) and testing methods have been developed for the major metabolite of (-)-Δ9-THC, (-)-11-nor-9-Carboxy-Δ9-THC glucuronide. The first CRM offered for this testing application was (+)-11-nor-9-Carboxy-Δ9-THC glucuronide which is a diastereomer of stereochemistry (6aS,10aS) of what is expected to be the naturally occurring metabolite. Studies have demonstrated that the major glucuronide metabolite identified in cannabis users differs from available reference material. Here we discuss the synthesis, certification, and confirmation of the first reference material for (-)-trans-11-nor-9-Carboxy-Δ9-THC beta-D-glucuronide which matches the major glucuronide metabolite observed in individuals following the use of cannabis.

Methods
(-)-trans-11-nor-9-Carboxy-Δ9-THC beta-D-glucuronide was synthesized via a multi-step sequence from chirally pure (-)-trans-THC intermediates. Multiple reaction monitoring (MRM) liquid chromatography tandem mass spectrometry (LC-MS/MS) was used to evaluate the chromatographic retention of (-)-11-nor-9-Carboxy-Δ9-THC glucuronide in whole blood specimens obtained from individuals which recently smoked cannabis and the synthetic reference.

Results
Synthesized (-)-trans-11-nor-9-Carboxy-Δ9-THC beta-D-glucuronide was certified using an orthogonal testing approach that results in a mass balance purity factor that accounts for chromatographic purity, water, residual solvent, inorganic content, etc. These tests are used in conjunction with structure identification techniques like NMR and LC-MS. MRM LC-MS/MS was then utilized to compare in vivo samples with the newly synthesized (-)-trans metabolite and available CRM, Cerilliant Cat # T-038 (+)-11-nor-9-Carboxy-Δ9-THC glucuronide. This comparison confirmed a difference in the observed retention times of T-038 with the new (-)-trans metabolite and the in vivo metabolite, while the latter two were observed to have matching retention times. Hydrolysis of both the in vivo metabolite and reference materials (T-038 and newly synthesized (-)-trans metabolite) resulted in a non-glucuronidated metabolite with matching retention times. This was an expected outcome as the hydrolyzed product of T-038 is the enantiomer of the synthesized (-)-trans and in vivo metabolite, and would not separate under achiral chromatographic conditions.

Conclusion
(-)-trans-11-nor-9-Carboxy-Δ9-THC beta-D-glucuronide was synthesized and manufactured as a CRM according to ISO 17034 for use in forensic and clinical LC-MS/MS applications. This is the first report that positively identifies the major glucuronidated form of THC as (-)-trans-11-nor-9-Carboxy-Δ9-THC beta-D-glucuronide in humans. The synthesized metabolite was compared against patient samples and chromatographically matched the in vivo metabolite observed in whole blood specimens. (-)-trans-11-nor-9-Carboxy-Δ9-THC beta-D-glucuronide is currently being developed into a solution standard under Cerilliant Cat # T-148-1ML and will be available in conjunction with the existing CRM, T-038. The availability of these reference materials provides the clinical and forensic fields with authentic reference materials for LC-MS/MS assays for the detection of (-)-11-nor-9-Carboxy-Δ9-THC glucuronide in clinical and forensic specimens.

Poster Presentations for

Nonspecific Adsorption and Loss of 11-Nor-9-carboxy-Δ9-tetrahydrocannabinol. Dude, where’s my THC? Triniti Jensen (Presenter) ARUP

Poster #27a View Map

This poster will be presented and discussed on Tuesday at 20:30 for 15 minutes in Montreal 3 (Track 2).

1. Problem

Here, we present the troubleshooting and lessons learned during the redevelopment of an assay to quantitate 11-Nor-9-carboxy-Δ9-tetrahydrocannabinol (THCA) in human urine by liquid chromatography tandem mass spectroscopy (LC-MS/MS). The initial development of the test following a traditional assay development process by a single scientist and the subsequent implementation into production to be performed by multiple staff members proved unsuccessful. The redevelopment of the test was then performed in close collaboration with the production lab. The second development was successful and produced a test with improved analytical performance, robustness and additional high throughput automated solutions to meet the increasing demand of this test.

2. Final Method

• Calibrators (CAL) and controls (QC) are prepared in authentic urine using a glass syringe
• 200 µL of CAL, QC, and patient samples are transferred to a Phenomenex glass-lined 96 well plate
• 400 µL of 5% KOH is added to samples for hydrolysis at 55°C for 30 min
• 250 µL 17% glacial acidic acid in methanol is added to quench
• A SPEware automated liquid dispenser (ALD) is used to automate sample clean-up coupled with a Phenomenex Strata-X-Drug B solid phase extraction (SPE) plate
• Final eluates are collected in a glass lined 96 well plate and diluted to LC starting conditions in 50:50 H2O:ACN
• Separation is performed with a Phenomenex Kinetex®2.6 µm biphenyl 100 Å, 30 x 2.1 mm column using 0.1% acetic acid in H2O and ACN with a total run time of 3.5 minutes on an Agilent LC system with a CTC-PAL autosampler.
• MS analysis is conducted on an AB SCIEX 5500 in negative mode

3. Troubleshooting Steps

• Recovery of THCA in synthetic and authentic urine:
o Average synthetic urine recovered 61% vs. 85% in authentic urine

Strategy 1: prepare CAL and QC in authentic urine
• Stability and solubility of THCA prepared in synthetic urine and stored at -80°C:
o After 1 and 5 months there was a loss of 33% and 55% of THCA, respectively

Strategy 2: CAL and QC cannot be stored in synthetic urine

Strategy 3: CAL and QC should be made fresh daily
• Solubility of THCA in extracted samples:
o Samples in 50:50 H2O:MeOH took more than 24 hours to equilibrate when stored at 4°C (constant increase in THCA area counts over time)
o Samples in 50:50 H2O:ACN were stable for 3 days stored at 4°C

Strategy 4: extracted samples need to be in 50:50 H2O:ACN
• Adsorption effects of plastic vs glass lined collection plates:
o Area counts of THCA-d3 were on average 50% lower when prepared in plastic vs glass plates
o Average accuracy of proficiency/spiked samples increased from 66% to 94%

Strategy 5: glass lined plates should be used during hydrolysis and collection steps
• Accuracy of measuring free THCA vs THCA-glucuronide spiked into authentic urine:
o Average accuracy of THCA (N=116) vs THCA-glucuronide (N=114) spiked into authentic urine was 90% vs 94%, respectively

Strategy 6: THCA-glucuronide spikes should be used to determine accuracy and precision
• Implementation of a less traditional “DevOps” approach during development with close collaboration between R&D and production staff coupled with the implementation of automation increased the robustness of the method:
o A combination of manual vs automated extractions was compared
o Six different bench staff preformed the assay to show it was reproducible

Strategy 7: When redevelopment is performed in collaboration with the production lab staff using lab resources assay robustness increases

4. Outcome

After the implementation of the above 7 strategies, a reproducible, robust, precise, and accurate assay was developed and successfully implemented into the production lab involving multiple staff members and shifts. Four quality control (QC) samples were monitored pre and post-test implementation. Imprecision of QC pre implementation was ±15% (N=141) versus ±5% (N=141) after implementation of the final method.

Interfering Peak in the Estradiol (E2U) LC-MS/MS Assay Mima Geere (Presenter) UCSF

>> POSTER (PDF)

>> ADDL INFO (PDF)

Poster #27b View Map

This poster will be presented and discussed on Tuesday at 20:45 for 15 minutes in Montreal 3 (Track 2).

Title: Interfering peak in the Estradiol (E2U) LC-MS/MS Assay

1. Problem

Estradiol is measured by a highly sensitive LC-MS/MS assay and utilized by clinicians to investigate different processes and disorders in women including menstrual cycle regulation, reproductive function and infertility as well as amenorrhea and oligomenorrhea. It is used to evaluate pregnancy maintenance, precocious puberty, menopause and estrogen deficiency in men and women.

There is an interfering peak in the LC-MS/MS results that requires further evaluation and separation. It is unclear if this peak is due to underlying patient medications or disease characteristics. The interference peak only occurs in 1% of sample runs in a select group of 7 patients in multiple samples collected over the last 7 months.

2. Method Information

Estradiol is measured using ultra-fast liquid chromatography (UFLC) coupled with tandem mass spectrometry (MS/MS). Estradiol is extracted from serum using a mixture of hexane and ethyl acetate, dried under nitrogen and derivatized using dansyl chloride. The sample is injected into the LC-MS/MS system where it is eluted onto the analytical column with a gradient of water and organic solvent (methanol and acetonitrile). Addition of an internal standard (carbon 13 labeled estradiol) allows for quantification of estradiol compared with a six-point standard curve and three quality control samples are run in each batch.

Instruments: Sciex 5500 QTrap mass spectrometer and Shimadzu Prominence UFLC
Column: Phenomenex Kinetex Phenyl-Hexyl, 2.6 micron, 100x3mm
Mobile Phase A: 0.1% formic acid in water
Mobile Phase B: 0.1% formic acid in 70% methanol and 30% acetonitrile
Injection volume: 30 microL
Column Temperature: 40 degrees Celcius
Autosampler temperature: 15 degrees Celcius

3. Troubleshooting Steps

Variations to increase Selectivity:

Modify the LC gradient time program going across the run over 8 mins.
Current time program:

Time (minutes) Flow rate (mL/min) % mobile phase B
1.0 0.8 20
1.2 0.8 70
4.5 0.8 95
5.4 0.8 95
5.5 0.8 20
6.5 0.8 stop

Variations in Troubleshooting:

1. Change the Gradient (timing in the program) – try different variations as follows
a. Change the timing of the gradient – Change to a longer or more complex gradient.
b. Change the organic composition of the mobile phase

2. Mass spec option: Collect product ion spectra of estradiol and of the interfering peak using the QTrap with a collision energy spread. This might tell us if there are any significant fragments that differ between the two peaks.

3. Change the columns from pheylhexyl to C18

4. Outcome
Troubleshooting experimentation in process, to be determined.

A Stumbling Block of Harmonizing LC-MS/MS Assays in Clinical Laboratories Hsuan-Chieh (Joyce) Liao (Presenter) University of Washington

Poster #28a View Map

This poster will be presented and discussed on Tuesday at 20:00 for 15 minutes in Montreal 3 (Track 2).

Problem:
One of the potential benefits of using LC-MS/MS instead of commercially available immunoassays is the ability to generate robust calibration materials de novo in-house and for oversight of the production process and quality assurance to be maintained within the laboratory. Upon calibration with matrix-matched calibrators, the concentrations of tocopherol calculated from the peak area ratio (ratio of unlabeled analyte peak area divided by the isotope-labeled internal standard peak area) in patient samples were different for the same extractions run on two identical LC-MS/MS systems. Using a standard method comparison, the slope of linear regression of the results from one instrument onto the results of the other instrument was at least 10-15% different.

Method Information:
• Serum extracted with sodium hydroxide and n-heptane (liquid-liquid extraction)
• Organic supernatant dried and reconstituted
• Acquity UPLC I-Class with a Column Manager
• Waters TQ-XS Tandem Mass Spectrometer
• Waters Acquity UPLC BEH C18, 2.1 x 50 mm
• Mobile Phase A: 2mM ammonium acetate, 0.1% Formic Acid in LCMS grade water
• Mobile Phase B: 2mM ammonium acetate, 0.1% Formic Acid in LCMS grade Methanol
• 5 mins gradient with 92% to 100% Mobile Phase B
• Quantitative MRM acquisition

Troubleshooting steps:
We assumed that a problem with the mass spectrometer hardware or software settings was most likely. The analytes were individually optimized on each instrument. We verified that gradients, transitions, cone voltages, and collision energies were all identical on the two instruments. The internal standard area was two- to five-fold higher on one of the instruments, which had a different probe type. One instrument was calibrated by the service engineer using a different calibration solution than what the laboratory staff used to calibrate the instrument (Orthophosphoric acid 0.1% Acetonitrile/Water versus Waters TQ-XS set up solution, which includes eight components in a 50% Acetonitrile/Water). Service was called for preventive maintenance.

Outcome
I wrote this up and looking for advice from other experts.
• Both instruments are almost brand new and maintain clinical specifications.
• Set up solution was ordered and the instrument will be re-calibrated using the correct unit mass calibration.
• Both instruments will be switched to the same style probe at the earliest convenience.
• Standards and controls will be run on the two instruments. Different collision energies would be performed to evaluate individual collision cells.
We have the same issues for the other analyte on the other two identical instruments. We think it is a general issue in every clinical lab but somehow it has not been fully addressed.

Morphine and Oxycodone Co-Positivity in Pain Management Urine Drug Testing Stephen Roper (Presenter) Washington University School of Medicine

>> POSTER (PDF)

Poster #28b View Map

This poster will be presented and discussed on Tuesday at 20:15 for 15 minutes in Montreal 3 (Track 2).

Our in-house pain management urine drug test is composed of a qualitative LC-MS/MS only opioid profile with low cutoffs for positivity combined with a standard screen and confirm approach for drugs of abuse. In rare cases, individuals with relatively large quantities of oxycodone in their urine have a small quantity of morphine that is concurrently detected. Discussion with pain management providers suggest that these individuals have a history compliance with prescribed medications and abstinence from non-prescribed substances. Review of method settings and results for contrived specimens does not suggest an analytical problem. Send out testing of these specimens to a referral lab that performs LC-MS/MS opioid testing using the same cutoffs for positivity typically returns an “interfering substance” comment for morphine results. This problem is complicated by the fact that some individuals at our institution are co-prescribed morphine and oxycodone, whereas others are only prescribed oxycodone only. Review of literature suggests that minor (unmonitored) metabolites of oxycodone may undergo in-source fragmentation to produce morphine isobars.

1.Problem:
Rare cases of co-positivity for morphine and oxycodone in individuals taking oxycodone only.

2.Method information:
•Dilute and shoot, 5μL injection
•Waters Acquity UPLC, 2.1x150 mm, 1.8μM C18 column
•Waters Xevo TQD, MRM acquisition, qualitative reporting
•8 minute linear gradient (0.4mL/min flow rate, 50°C column temp)
•Opioid Cutoffs for positivity:
•25 ng/mL: Morphine, hydrocodone, hydromorphone, oxycodone, oxymorphone, tramadol, O-desmethyltramadol, tapentadol, Methadone, EDDP
•10 ng/mL: 6-acetylmorphine and naloxone
•5 ng/mL: Buprenorphine and Norbuprenorphine
•All other drug classes tested by standard screen and confirm approach

3.Troubleshooting:
•Review of method settings and analysis of contrived specimens did not correct the problem. Nor did further separation of morphine and oxycodone acquisitions in time. Investigation suggests morphine is most likely real. While it may be present for several reasons, co-positivity with oxycodone may indicate in-source fragmentation of a minor metabolite of oxycodone that results in production of a morphine isobar [1].

4.Outcome:
•Currently, any pain management specimen that is co-positive for oxycodone and morphine requires a positive opiate immunoassay result in order to report morphine.

[1] A.C. Muñoz-Muñoz, T. Pekol, D. Schubring, C. Johnson, L. Andrade, Identification of Novel Opioid Interferences using High-Resolution Mass Spectrometry, J Anal Toxicol 42(1) (2018) 6-16.

Interfering Compound in Urine Cannabinoid Analysis Agnes Cua (Presenter) Precision Diagnostics

Poster #29a View Map

This poster will be presented and discussed on Wednesday at 17:45 for 15 minutes in Montreal 3 (Track 2).

PROBLEM

An interfering compound coelutes with delta-9-THC-COOH (11-Nor-9-carboxy-delta-9-tetrahydrocannabinol) in urine drug analysis, with a LC-MS/MS method that uses Phenyl Hexyl column and a 4.2 min gradient program. This causes a small shift in the retention time (RT) of delta-9-THC-COOH preventing its identification, and subsequently is reported as negative or with interference comment, even when delta-9-THC-COOH is present in the urine sample. There is also a possibility that with the RT shift being very small, it becomes not significant, and peak is identified as delta-9-THC-COOH. This becomes an issue when delta-9-THC-COOH is absent in the urine sample and is reported as positive in the presence of the interfering compound.

We also sought to identify the identity of the interfering compound.

METHOD INFORMATION

50 µL urine processed with ‘Dilute and Shoot’ method
Shimadzu LC – MPX (2 streams)

A. SCIEX 6500(+) Mass Spectrometer
MP-A: 0.1% Formic Acid in H2O
MP-B: Methanol
4.2 min gradient @ 0.8-1.2 mL/min flow rate
Phenyl Hexyl column, 50 x 4.6 mm, 2.6 µm, with guard cartridge
Column oven @ 50oC
5 µL injection volume
Quantitative MRM acquisition:
delta-9-THC-COOH 345: 193, 119
delta-9-THC-COOH-d3 348: 302


TROUBLESHOOTING STEPS

To resolve the issue, a LC-MS/MS method that uses a C18 column was developed, where an initial isocratic phase with 60% B was followed by an increase to 95% B and subsequent re-equilibration.

B. SCIEX 5500 Mass Spectrometer
MP-A: 2mM Ammonium Acetate in H2O
MP-B: Methanol
3.5 min isocratic/gradient, 0.8 mL/min flow rate
C18 column, 50 x 2.1 mm, 5 µm, with guard cartridge
Column Oven @ 40oC
5 µL injection volume
Quantitative MRM acquisition:
delta-9-THC-COOH 345: 193, 119
delta-8-THC-COOH 345.2: 299.1, 165.1
delta-9-THC-COOH-d3 348: 302

For identification of interfering compound, a delta-8-THC-COOH (11-Nor-9-carboxy-delta-8-Tetrahydrocannabinol) Standard was used to confirm its identity.

OUTCOME

The method results in the separation of the interfering compound from delta-9-THC-COOH, allowing definite quantitation of delta-9-THC-COOH. Used of the Standard also allowed identification of the interfering compound to be delta-8-THC-COOH.

Communicating Proteogenomics Data to Physicians John Koomen (Presenter) Moffitt Cancer Center

Poster #29b View Map

This poster will be presented and discussed on Wednesday at 18:00 for 15 minutes in Montreal 3 (Track 2).

Summary

Molecular diagnostics for lung cancer have become increasingly complex. Multiplexed genomics assays can be complemented with the proteomic phenotypes to further support clinical decision-making, yet these increases in data produce additional barriers to the effective use of this information in the clinic. While individual lab test results can be communicated effectively from the clinical lab to the patient and physician, cancer researchers and treating physicians are now faced with a large variety of therapeutic options that have specific companion diagnostics. In order to make the information available and to support ranking of therapeutic options to maximize expected patient benefit, complex proteogenomic data must be related to the physicians in an accessible way that still enables thorough review.

Abstract Detail

1. Problem
Detailed molecular characterization of lung cancer is necessary to rank therapies according to their expected benefit to the patient. Numerous treatment options are now available, including chemotherapy, targeted therapy, and immunotherapy as well as novel approaches in clinical trials. In addition, some tumor subtypes (e.g. adenocarcinomas driven by ALK fusions) are observed only in a small proportion of the patient population. Targeted proteomics and genomics panels including drug targets and known biomarkers can be used to characterize genes and proteins and help direct treatment for each individual cancer patient. However, the results often indicate multiple options could be effective, so better methods are needed to triage patients for FDA-approved companion diagnostics, specific treatment regimens, and clinical trials.

2. Method Information
We are developing an application that would enable the user to review the proteogenomic profile of each individual tumor in the context of the overall population for patient classification. In addition, data for patient outcomes will also be included to assist the physician in prioritizing regimens that would provide the most benefit based on the molecular characterization from proteogenomics. Data will be presented to summarize genomic and proteomic data for each gene/protein. Heat maps with unsupervised clustering visualize molecular classification of tumors. Waterfall plots displaying the levels of protein or gene expression can rank the patient in the context of the population. Ultimately, these data can be linked to outcomes and used to select therapies for each individual patient.

3. Troubleshooting Steps
A test case has been developed using lung cancer cell lines that leverages publicly available genomics data and drug sensitivities as well as proteomics data generated in-house to explore data visualization. Data acquisition is underway for tumor tissue samples using different clinical vignettes (e.g. response to EGFR or MET targeted therapy) to help further develop the ability to communicate complex data acquired from patient specimens.

4. Outcome
The goal for this troubleshooting poster is to obtain feedback on our communication strategy from the MSACL community. Comments on the existing strategies and suggestions for additional novel approaches are welcome.