Abstracts for the website 6-22

Abstract 137
Clinical Sample Treatment, Absolute Matrix Effects and Chromatography: A Look at the Impact
on Bioanalytical Results
Ryan W. Lutz, Cynthia M. Chavez-Eng, Marvin L. Constanzer and Eric J. Woolf,
Merck Research Laboratories, West Point, PA
Objectives: Compound I, investigated for treatment of Type 2 Diabetes, required sample treatment with
phosphoric acid to stabilize a potential glucuronide metabolite observed in animal species. During
sample analysis for the first clinical trials, the sample treatment procedure and analytical protocol needed
to be modified to minimize variable response due to sample pH and absolute matrix effects. Methods:
Analyte and stable labeled internal standard were isolated from acidified human plasma via liquid-liquid
extraction. Samples were quantified by HPLC-MS/MS using electrospray ionization under negative
polarity. Initial chromatographic conditions utilized a C18 column with a polar embedded phase.
Clinical sample analysis results using this method revealed a highly variable internal standard response.
Results: Clinical protocol procedure was modified to ensure consistent treatment of plasma samples, as
recovery was highly dependent on sample pH. The analytical protocol was modified to minimize the
absolute matrix effect by utilizing the atmospheric pressure chemical ionization source and by changing
the chromatographic conditions such that absolute matrix effect was minimized. Implications: Potential
issues with validated bioanalytical methods often do not take form until the first in man and subsequent
clinical studies. It is important to consider all aspects of bioanalysis, including sample treatment at the
clinic, in order to isolate and solve these issues. Collection procedure, ionization source, and
chromatography were all necessary points of investigation in order to minimize variable response.
Abstract 138
Internal Standard Response Variation during Incurred Sample Analysis by LC-MS/MS: How
Much Is Too Much?

Aimin Tan, Adrien Musuku, and Robert Massé, Anapharm, Ste-Foy, Québec, Canada
Objectives: Internal standard (IS) responses can directly impact the accuracy of reported concentrations
in bioanalysis as most methods are based on analyte/IS response ratios for quantitation. Due to the
complexity of incurred sample matrices, variable IS response is quite common upon applying a validated
method to the analysis of incurred samples. But how much variation in IS response is acceptable?
Methods: Firstly, different real cases of IS response variations observed during the analysis of incurred
samples were investigated for root-causes. Then, the impact of these variations on the accuracy of
reported concentrations was evaluated. Results: Twelve cases were investigated, such as error in addition
of IS, random and sharp drop in IS response, gradual decrease/increase of IS response, high/low IS
response for incurred samples only, less IS response variation observed but accuracy was affected,
randomly scattered low IS responses for incurred samples only and not repeated during re-analysis,
deuterated IS not following the analyte and reinjection results not matching those of 1st injection.
Implications: IS response variation could be caused by many different factors and therefore, each case
should be dealt individually. Neither no criterion at all nor a set cut-off criterion is appropriate. Ideally,
the difference in IS response between an incurred sample and calibration standards and quality controls
should be evaluated first to determine whether it was a random occurrence or there were patterns/trend.
Accordingly, the identified samples could be reanalyzed or the results could be accepted with an
investigation to demonstrate that the accuracy was not affected.
Abstract 139
Identifying And Overcoming Bioanalytical Challenges Associated With Chlorine-Containing
Dehydrogenation Metabolites.

Michael T. Furlong1, Chad E. Wujcik2 and Yi Su1; 1Pharmacokinetics Dynamics and Metabolism, Pfizer
Global Research and Development, Groton CT, 2Monsanto Corporation, St. Louis, MO
Objectives: Assess strategies to minimize quantitative interference of co-eluting dehydrogenation meta-
bolites during the bioanalysis of chlorinated small molecules. Methods: LC-MS/MS experiments were
conducted using Shimadzu pumps and a CTC Analytics HTS PAL autosampler, coupled to an Applied
Biosystems API 4000 tandem quadrupole mass spectrometer equipped with a TurboIonspray™ source.
Reverse phase chromatography was performed on an ACE 3 C18 column. Plasma samples were extract-
ed by acetonitrile precipitation. Results: During LC-ESI-MS/MS analysis of plasma samples from a rat
exploratory toxicology study two unexpected interfering peaks were observed to nearly co-elute with the
chlorine-containing analyte. Q1 precursor ion scans of study samples indicated that these peaks contained
putative [M-2] (loss of 2 Dalton) metabolites. Strong apparent isotopic contributions at [M+2] were noted
in both cases, suggesting that the putative metabolites retained the chlorine atom of the parent compound.
The potential isotope-related impact of these metabolites on quantification of the parent compound was
assessed. Several alternate precursor ion and product ion combinations were evaluated and shown to
minimize the quantitative impact of the interfering metabolites without having to rely on their chromato-
graphic resolution from the parent compound. Implications: These results indicate that when quantifying
chlorine- or bromine-containing small molecules from in vivo samples or in vitro metabolic incubations,
1) efforts to detect potential dehydrogenation metabolites should be undertaken, and 2) if such meta-
bolites are detected, the judicious choice of alternate multiple-reaction monitoring (MRM) transitions can
limit their impact on quantification of the parent molecule without the need for robust chromatographic
resolution.
Abstract 140
Quantitative Interference by Cysteine and N-Acetylcysteine Conjugates During the LC-MS/MS
Bioanalysis of an Electrophilic Small Molecule
Michael T. Furlong, Jason Barricklow and Tim F. Ryder; Pharmacokinetics Dynamics and Metabolism,
Pfizer Global Research and Development, Groton CT
Objectives: Identify interfering metabolites discovered during bioanalysis of a small molecule in human
urine samples, assess their quantitative impact, and develop strategies to minimize the interference.
Methods: Quantitative LC-MS/MS experiments were conducted using Shimadzu pumps and a CTC
Analytics autosampler, coupled to an Applied Biosystems API4000 triple quadrupole mass spectrometer
equipped with a heated nebulizer probe or a TurboIonspray™ probe. Metabolite identification experi-
ments were carried out on both Orbitrap and TSQ systems (Thermo). Reverse phase chromatography was
performed on Phenomenex Onyx C18 columns.Urine samples were extracted by acetonitrile precipitation.
Results: During LC-APCI-MS/MS analysis of a small molecule in urine samples from a human pharma-
cokinetic study, an unexpected interfering peak was observed to nearly co-elute with the analyte.
Chromatographic separation revealed the presence of at least 3 metabolites, which were identified as two
diastereomeric cysteine conjugates of the analyte, as well as their N-acetylated (mercapturic acid)
derivatives. Quantitative comparison using the original chromatographic method and the improved
separation method indicated that the original method substantially over-estimated the analyte
concentration in many study samples. The quantitative over-estimation in the original method was shown
to be due to in source fragmentation of the metabolites to the analyte. Implications: These results 1)
demonstrate the potential for compromised quantification of electrophilic small molecules in biological
matrices due to co-eluting thiol metabolites, 2) underscore the need to carefully evaluate LC-MS/MS
methods for electrophilic small molecules to ensure that they are not susceptible to such interferences
during quantification of in vivo samples.
Abstract 141
Development and Validation of a LC-MS/MS Method for the Determination of Salmeterol in
Human Plasma
John P. Allanson, Enric Bertran, Stephen D. Clarke, Irene Morelli, Terry A.G. Noctor, Helen Young,
Unilabs York Bioanalytical Solutions, York, UK
Objective: Salmeterol is a long-acting β2-adrenergic receptor agonist drug that is currently prescribed for
the treatment of asthma and chronic obstructive pulmonary disease. Published methods for the
determination of salmeterol are not sufficiently sensitive to measure levels typically present at the later
phases of elimination. A more sensitive method (lower limit of quantitation 1 pg/mL) was therefore
developed. Methods: A solid phase extraction (SPE) method was developed for the extraction of
salmeterol from plasma samples (250µL). The method employed a polymeric mixed-mode cation
exchange medium. Salmeterol-d3 was used as internal standard. HPLC separation was achieved an ODS
silica column using a mobile phase gradient comprising a formic acid - methanol mixture, delivered at a
flow rate of 400 µL/minute. Tandem mass spectrometric detection was performed using an MDS Sciex
API 5000 instrument using a Turbo IonSpray interface (positive ion mode). Results: The method was
validated over a range of 1 to 200 pg/mL. Inter-run precision was ≤9.5% while bias ranged from -2.5% to
8.0%. The method was determined to be specific for the determination of salmeterol and free from matrix
effects investigated in six different individual plasma samples and in lipaemic and haemolysed plasma.
The method does not suffer from interference from co-administered fluticasone propionate. Implications:
A sensitive, accurate and robust method for the determination of salmeterol in human plasma was
developed, validated and successfully applied for the quantitation of salmeterol in clinical samples.

Abstract 142
Process Optimization of Bioanalytical Support for Merck Development Compounds; Leveraging
One Analytical Assay to Generate Data for Multiple Analytes and Multiple Matrices.
Ashley N. Martucci, Michelle Groff, Cynthia Miller-Stein, Department of Drug Metabolism and
Pharmacokinetics, Merck Research Laboratories, West Point, PA
Objective: A semi-automated liquid-liquid extraction method was developed and validated to quantitate a
Merck compound (Compound A) in human plasma to support pharmacokinetic studies. This assay was
then applied to determine levels of drug in human urine and levels of metabolite in human plasma with
minimal additional development or validation effort. Method: The analytical method was automated to
extract 300 µL of plasma samples with 0.8 mL ethyl acetate in a 96-well plate. Ethyl acetate was
evaporated and samples were reconstituted with 0.150 mL of 45:55 methanol:water. Compound A and
it's hydroxylated metabolite were separated by reversed phase HPLC isocratically with a mobile phase of
55:45 methanol:0.1% formic acid. Tandem mass spectrometry in Multiple Reaction Monitoring mode
with a turbo ionspray interface in positive ion mode was used for detection. Compound A had a dynamic
range of 0.2-200 nM. Metabolite ranged from 1-1000 nM. Control urine and urine samples were diluted
100 fold with control plasma and analyzed using the validated plasma assay. Results: For the metabolite,
three standard curves were extracted and analyzed. Intraday precision varied from 3.23%-7.00% and
accuracy ranged from 94.64%-107.14%. Recovery of metabolite in plasma was 94.15%. For the urine
assay, replicate QC's (n=5) were analyzed. Precision was 1.54%, accuracy ranged from 96.98%-100.68%,
and recovery was 98%. Implication: By leveraging one analytical assay across various matrices and
analytes, method development time and resources were saved. The data generated was then used to
determine the necessity of further developing metabolite and urine assays.
Abstract 143
A Sensitive LC-MS/MS Method for the Quantitation of Fluticasone Propionate in Human Plasma
John P. Allanson, Enric Bertran, Stephen D. Clarke, Irene Morelli, Terry A.G. Noctor, Karen Tennant,
Unilabs York Bioanalytical Solutions, York, UK
Objective: Fluticasone propionate (FP) is a potent, synthetic corticosteroid used to treat asthma and
allergic rhinitis. Literature methods for FP in plasma have lower limits of quantitation which are often
too high (20 - 50 pg/mL) to allow quantitation of FP in plasma more than a few hours after administration.
The objective of this work was therefore to develop and validate a sensitive LC-MS/MS method for the
determination of FP in human plasma, with a lower limit of quantitation of 3 pg/mL. Methods: A solid
phase extraction (SPE) method utilising a C18-modified silica sorbent was developed (sample size 500
µL). FP-d3 was used as an internal standard. HPLC: ODS silica column; mobile phase gradient
comprising ammonium formate (25 mM, pH 5) - methanol, 300 µL/minute. Tandem mass spectrometry:
MDS Sciex API 5000 instrument, TurboIonSpray (positive ion mode). Results: The method was
validated over a range of 3 to 200 pg/mL. Inter-run precision was ≤10.0% while bias ranged between -
1.0% to 15.0%. The method was demonstrated to be specific for the determination of FP, free from
matrix-related modification of ionisation effects investigated in six different individual plasma samples
and to be unaffected by lipaemia and haemolysis. Implications: A sensitive, accurate and robust method
for the determination of FP in human plasma over the range 3 - 200 pg/mL was developed, validated and
successfully applied to the quantitation of FP in clinical samples. The method is suitable for the detection
of FP in samples at least 24 hours after an inhaled dose of 100 µg.

Abstract 144
Development and Validation of an LC-MS-MS Method for the Quantitation of Anastrozole in
Human Plasma (K2EDTA)

Ted J. VanNoord1, Heather M. Dotzauer1, Timothy J. Schuchardt1, Nicole Hudson1, Patricia A. Hansen1,
Michael J. Herrera1, Sandra M. Sims2, Steven J. Weber1, 1Eurofins AvTech Laboratories, Inc., Portage,
MI, 2Meditrina Pharmaceuticals, Inc., Ann Arbor, MI
Objective: The objective for this project was to develop and validate a method for quantitation of
anastrozole in human plasma. Method: Anastrozole and an internal standard (letrozole) were extracted
from 100 µL of human plasma using 1.0 ml of methyl tert-butyl ether. Following a flash freeze in a dry
ice-alcohol slurry, the organic layer was poured into a second tube and evaporated. Samples were
reconstituted in 1.0 mL of water:acetonitrile, 70:30, v:v and 10 µL of each sample was injected onto an
LC-MS system. The analytes were eluted from a Waters Symmetry C18 column (3.5 µm, 2.1 x 50 mm)
with a mobile phase of 0.1 % glacial acetic acid in water:acetonitrile, 68:32, v:v, followed by a column
wash of 0.1 % glacial acetic acid in water:acetonitrile, 20:80, v:v between each injection. Detection was
performed using an Applied Biosystems API 4000 using electrospray ionization in the positive ion mode.
Results: The method was validated over the range of 2.00 to 500 ng/mL. Inter-batch accuracy (%RE) for
quality control samples of anastrozole ranged from 0 to 4%, while inter-batch precision (%RSD) ranged
from 6 to 8%. Acceptable accuracy and precision was shown at the LLOQ level. Specificity was shown in
six lots of plasma by infusion experiments and ion suppression experiments. Short-term stability in
plasma at ambient conditions and following three freeze-thaw cycles was also shown. Conclusion: A
sensitive, accurate, and reproducible method was developed and validated for the analysis of anastrozole
in human plasma and was applied to clinical sample analysis.
Abstract 145
Analysis of Drug Metabolites in Biological Fluids using Mixed-Mode Solid Phase Extraction and
UltraPerformance Liquid Chromatography-Tandem Mass Spectrometry

Zhe Yin, Kenneth J. Fountain, Erin E. Chambers, Paul D. Rainville, Diane M. Diehl
Waters Corporation, Milford, MA, USA
Objective: The development of a bioanalytical method for the determination of morphine and
compounds. Methods: A method using mixed-mode solid phase extraction and Ultra Performance liquid
chromatography-tandem mass spectrometry was developed for the analysis of morphine, morphine-3β-
glucuronide, morphine-6β-glucuronide, 6-acetylmorphine, morphine N-oxide, and 10-hydroxymorphine
in porcine plasma. All six compounds, along with four deuterated internal standards, were
simultaneously extracted using mixed-mode strong cation exchange SPE in a 96-well µElution plate
format. Due to analyte instability, a neutralizing solvent was used during the elution step to minimize
degradation of 6-acetylmorphine. Separation was performed on a 2.1 × 100 mm, 1.8 µm C18 column
designed for retention of extremely polar compounds using a formic acid and methanol gradient.
Analytes were detected by positive electrospray ionization in multiple reaction monitoring mode. Results:
Recovery was 77-120 % depending on the analyte, and inter-day variability was less than 6 %. Linearity
was determined in porcine plasma by spiking the analytes prior to SPE. Correlation coefficients were ≥
0.998, and % deviation from the actual concentrations was less than 15 %. The lower limit of
quantitation for all compounds was between 0.1 and 0.25 ng/mL. Implications: A method was
developed to simultaneously extract and preserve analyte structural integrity during sample preparation
and subsequent quantification of parent drug and associated polar compounds.
Abstract 146
Dextromethorphan Metabolism by Human and Rat Liver Microsomes by UPLC QTOF Accurate
Mass MS/MS with Isotopically-Labeled Parent and Metabolite Standards.

Mark J. Horrigan, Joanna E. Barbara, Paul Toren and Andrew Parkinson, XenoTech, LLC, Lenexa,
Kansas
Objectives: To investigate the in vitro metabolism of dextromethorphan by human and rat liver
microsomes by UPLC QTOF MS/MS analysis with and without stable-isotope labels for comprehensive
metabolite structural elucidation. Methods: Dextromethorphan and major metabolites, native and
deuterated, were incubated with UDPGA- and NADPH-fortified liver microsomes from human and rat.
Samples were analyzed by UPLC and tandem accurate mass spectrometry with electrospray ionization.
Metabolite profiling and characterization for all incubated compounds were performed. Results: In
addition to the expected major metabolites, several unexpected hydroxylation metabolites of
dextromethorphan were detected. Comprehensive fragmentation pathways for dextromethorphan and
metabolites were determined from accurate mass data acquired for the native and stable isotope-labeled
compounds. Comparison of fragmentation data acquired for the native and labeled compounds facilitated
enhanced structural elucidation of the metabolites. Implications: Dextromethorphan is of interest as both
a widely available over-the-counter drug and as an FDA-approved in vivo and in vitro probe substrate for
the polymorphically-expressed cytochrome P450 enzyme CYP2D6. Profiling of its major metabolites has
been previously published, but metabolite structural elucidation has been limited by unusual MS
fragmentation behavior. Incubation of a mixture of stable isotope-labeled and unlabeled drug for
metabolite profiling has been widely reported for various test compounds, but separate incubation of
native and labeled drug and metabolite standards is not currently used to improve metabolite structural
elucidation. This study illustrates the value of such an approach.
Abstract 147
High Throughput Screening GSH Adducts Using Hybrid Linear Ion Trap Systems Coupling with
Fast chromatography at Clinically Relevant Dose Concentration

Elliott Jones, Claire Bramwell-German, Hesham Ghobarah, Hua-fen Liu and Tanya Gamble, Applied
Biosystems/MDS Analytical Technologies, Concord, Canada.
Objective: The detection of reactive metabolites such as GSH adducts has become a common practice in
early drug toxicity screening. In order to detect all potential GSH adducts, two or more injections were
often used. Higher substrate concentrations (10-50 uM) which may be higher than clinical relevant
concentration and complex sample preparation were often needed in order to achieve the sensitivity
desired. This paper investigates combining LC/MS/MS scan modes on a hybrid linear ion trap for high
throughput (HT) screening of GSH adducts at clinically relevant concentrations in a single injection.
Methods: UPLC-like conditions with an acquisition time of 5 to 10 minutes. Two MS information
dependent acquisition (IDA) methods were explored. The first combines neutral loss (+129, +307) and
precursor ion (-272) scan modes for survey scans. The second method uses the characteristic
fragmentation pattern of GSH to build predictive MRM (pMRM) transitions. Both methods make use of
fast polarity switching between scan modes. Results: Trazadone was used as a model compound to
investigate GSH formation at 2 and 20 µM in human liver microsomes. The two MS approaches were
able to detect GSH conjugates at trace levels not easily observed using standard LCMS workflows. The
pMRM approach proved to be more sensitive, but relies on the ability to predict the biotransformation.
Implications: A sensitive HT method for complete GSH detection and structural confirmation at
clinically relevant concentrations ensure better in vivo prediction.
Abstract 148
Development and Validation of a Simple, Cost Effective LC-MS/MS Method for the Quantification
of Mevalonic Acid (a Biomarker for Cholesterol Synthesis) in Human Urine.
Noel D.Premkumar, Hong Li, Jeffrey P. Brendler, Amy L. Mize, Alana L. Eppinger, Hui Wu, Analytical
Bio-Chemistry Laboratories, Inc.

Objective: To develop a simple, cost effective LC-MS/MS method for the quantification of Mevalonic
acid (MVA) in human urine. Methods: A liquid/liquid extraction technique was developed using tert-
butyl methyl ether (TBME). The calibration curve was prepared in water due to endogenous MVA
present in urine. The linear range was 10.0 - 2500 ng/mL. Mevalonolactone-1,2-13C2 was used for the
internal standard. Mevalonic acid lactone was spiked into human urine at three concentrations for quality
control samples. Samples were acidified to convert any MVA present to lactone form, and allowed to sit
for one hour at room temperature. TBME was added, the samples were vortexed and centrifuged. The
organic layer was collected and dried under nitrogen then reconstituted using a basic solution to convert
the lactone into the acid form for analysis. Separation was achieved on a Varian Monochrome C18
column (100 x 4.6, 5µm) at a 0.700 mL/min flow using a gradient. Analysis was done on an AB Sciex
4000 mass spectrometer using TurboIon spray. Detection was by negative ion mode with multiple
reaction monitoring. Results: The method was linear 1/x with correlation coefficients >0.9900. Intra-
run mean accuracies ranged from 93-107% over three separate days, with inter-run accuracies from 95-
104%. Precision ranged from 1.0-3.1 %CV. Implications: MVA is an important biomarker of
cholesterol and bile acid synthesis. This validated method is simple, cost effective, robust and repeatable,
thus making it an attractive method for evaluating changes of the rates of cholesterol synthesis post
cholesterol lowering treatments.
Abstract 149
Strategies to Increase the Throughput of Drug Quantitation in Tissues for Discovery Support

Robert Papp, Ria Seliniotakis, Kevin P. Bateman, and Laird A. Trimble, Merck Frosst, Montreal, Quebec,
Canada.
Objectives:
The goal of this work was to explore streamlined approaches to 1) homogenize several rodent tissue types such as lung, skin, trachea, bronchi, liver, kidney, brain, harderian glands and 2)
evaluate the success rate of a plasma standard curve as surrogate matrix for quantitating tissue drug levels.
Methods: Several commercial tissue homogenization devices were evaluated for their ease of use, speed
and ability to effectively homogenize tissues. Device types included bead beating, polytron and
ultrasonic systems. Tissue homogenates from liver, kidney and lung were analyzed using standards
prepared in a tissue matrix as well as standards prepared with a plasma matrix supplemented with controls
spiked into pooled tissue extracts. UHPLC-MS/MS was used for all quantitation. Results: The
FastprepTM tissue homogenization system showed superior efficiency at homogenizing the most difficult
tissues such as skin, trachea and bronchi. The GenogrinderTM in a 96well format showed the highest
throughput and best possibility for automation. 88% of the compounds tested could be quantitated with a
plasma standard curve with less than 30% deviation from the tissue curve results. The use of controls
spiked in pooled tissue extracts did not provide additional benefit. Implications: Increased throughput
and versatility is achieved by using the FastprepTM and GenogrinderTM tissue homogenization systems. A
reduction in use of laboratory and animal resources was demonstrated by using a plasma calibration curve
as a surrogate for tissues.

Abstract 150
Quantitation of Abbott-Compound In Biological Matrices By MALDI/MS/MS As Compared To
LC/MS/MS.

Azza M. Wagdy, Abbott Laboratories, Patrick Pribil, MDS Analytical Technologies, Joseph C. Kim,
Abbott Laboratories, Huaiqin X Wu, Abbott Laboratories ,and Tawakol A. El-Shourbagy, Abbott
Laboratories.
Objectives: A MALDI-MS/MS method has been developed for the quantitation of Abbott-Compound, a
small drug molecule. An investigation was conducted to evaluate MALDI FlashQuant as compared to a
validated LC/MS method for the analysis of biological samples. Methods: Abbott-Compound and its (d-
IS) in rat and human plasma were extracted using 96-well LLE with 1/1 ethyl acetate/hexane. For
MALDI, the dried extracts were dissolved into 50 µL α-cyano-hydroxycinnamic acid matrix, (6 mg/mL in
60% acetonitrile/40% water/0.1% TFA). 1µL sample was deposited in triplicate onto MALDI 384-plate,
analyzed at 1.5 mm/sec. Optimized MRM method with +/ions was used for both techniques. The rat (AP)
and human (HP) plasma STDs ranged from 0.025 to 2.03 µg/mL and 0.116 to 25.8 ng/mL respectively.
Matrix Effect (MEQCs) (21 ng/mL) in HP were assessed in six lots. Results: For AP, mean bias: LLOQ
2.8% with CV 11.1%, QCs between -0.4% and 6.4% and CV ≤ 11.0%. 80% of analyzed rat study
samples by MALDI matched the values by LC/MS with difference of +/- 20% and CV ≤ 16.8%. For HP,
mean bias: LLOQ -10.4% and 17.7% CV, QCs between 68.2% and 1.73% with CV ≤ 36.9%. MEQCs
spots, mean %bias between -1.16% and 1.14% of theory with CVs ≤ 3.4%. The MALDI run time was 22
minutes for 288 spots, and LC/MS was ~ six hours for 98 samples. Implications: MALDI was 20X
faster than LC/MS acquisition time and assay reproducibility, with acceptable accuracy and precision,
was in the range of 21 to 2000 ng/mL.
Abstract 151
Validation of a Chiral LC-MS/MS Assay for R- and S-Warfarin in Human Plasma to Support
Drug-Drug Interaction Studies

Abram Brubaker, Ly Tran, Bernard Ntsikoussalabongui, Jacqueline Killmer, and Dhruba Bhattarai, MPI
Research, Mattawan, MI.
Objective: Warfarin is used extensively as an oral anticoagulant for the prevention and treatment of
thromboembolic disease. Results from a study published in 2006 reveal that 14.7% of patients who
participated in the study experienced a hemorrhage event within one week after beginning warfarin
therapy. It has been well documented that these adverse events typically are the result of a
pharmacokinetic interaction between a concomitant drug and warfarin involving inhibition of cytochrome
P450 2C9. The FDA requires that new potential drugs must be evaluated for potential drug metabolism
and interactions with warfarin. A bioanalytical method was needed to separate concomitant drugs from
wafarin to minimize them as potential interefences in the detection of warfarin extracted from plasma.
Method: The chiral LC-MS/MS assay was validated using a Chirobiotic V™ analytical column. Wafarin
was extracted from plasma using a MCX 96-well plate to trap basic amine drugs on the extraction bed
while eluting warfaarin from the neutral phase. Results: A full validation of the chiral LC-MS/MS assay
for R- and S-warfarin in the concentration range of 5 to 2500 ng/mL was carried out successfully.
Implications: Having a bioanalytical method which separates basic drugs from warfarin during the
plasma sample extraction reduces the risk of these drugs as being potential interferences. In turn, this
greatly increases the ruggedness of the method.
Abstract 152
Validation of a Chiral LC-MS/MS Assay for R(+)- and S(-)-Bupivacaine in Human Plasma to
Support Bioequivalency Studies.

Jacqueline D. Killmer, Dhruba B. Bhattarai, Troy Frayer, Bernard Ntsikoussalabongui, and Abram N.
Brubaker, MPI Research, Mattawan, MI

Objective: The first objective was to validate a robust and rapid chiral bioanalytical assay for clinical
bioequivalency studies of bupivacaine using low sample volumes. The second objective was to probe the
mechanisms of chromatographic chiral recognition. Method: The chiral bioanalytical assay was
validated using supported liquid extraction and HPLC/MS/MS. The Chirobiotic-V™ column was
selected for method development studies due to its broad chiral recognition mechanisms. Results: A full
validation of the chiral bioanalytical method for the analysis of bupivacaine enantiomers within 14 min in
the range of 1-1000 ng/mL using a sample volume of 50 µL human plasma was completed successfully.
The affects of flow rate, organic strength, buffer strength, and pH on chiral recognition were evaluated for
optimization of enantiomeric resolution. Baseline resolution of 3.1 was achieved. The affect of the alkyl
chain length of the analyte was also evaluated using the series N-(2,6-dimethylphenyl)-2-
piperidinecarboxamide mepivacaine, ropivacaine, and bupivacaine. Implications: Bupivacaine is used
extensively in the clinic as a local anesthetic, nerve blocker, and for epidural and intrathecal
anesthesia. The cardiotoxic effects of local anesthetics are well documented and it was demonstrated that
bupivacaine's toxicity is due to the R(+) enantiomer. Future clinical bioequivalency studies will require
robust chiral assays such as that developed and validated by MPI research. The low volume and fast run
time of this chiral assay coupled with a thorough understanding of the mechanisms of chiral recognition
yield a method ideal for such studies.
Abstract 153
Identification of the in vitro Glucuronidation and Sulfation Pathways of ARQ 501 (β-lapachone)
Using Human UDP-Glucuronyltransferases and Sulfotransferases.

Karen R Bresciano, Terence Hall, Ronald E Savage, Robin S Goldstein, ArQule Inc.,
Woburn, MA.
Objective: ARQ 501 (β-lapachone) is an investigational compound for the treatment of advanced solid
tumors. The in vivo metabolism of ARQ 501 observed in humans, and also in human hepatocytes, is
primarily conjugation followed by biliary excretion. The identification of the specific UDP-
glucuronyltransferase (UGT) and sulfotransferase (SULT) forms responsible for conjugative
biotransformation of ARQ 501 is currently being explored. Methods: The in vitro assays are conducted
using a variety of commercially available enzymes. The assays are also conducted in the presence and
absence of potential activating compounds and enzymes to determine what role, if any, they have in the
metabolism of ARQ 501. Results: Our goal is to identify the specific isoforms of UGT's and SULT's
responsible for biotransformation of ARQ 501. Studies are currently ongoing, but preliminary data
suggests that one or more co-factors not normally required to initiate glucuronidation in vitro may be
necessary in the case of ARQ 501. Implications: Identifying the specific enzymes which metabolize
ARQ 501, and potentially any additional co-factors or components required, will allow an understanding
of the pathways involved in its metabolism and elimination. Additionally, it may shed light on the
metabolism of other compounds that share the same structural motif.
Abstract 154
Utilizing Fast LC-MS/MS for High-Throughput Bioanalytical Analysis In GLP Environment

Katty X. Wan, Jacob D. Sadik, Jill E. Polzin and Matthew J. Rieser; Abbott Laboratories, Abbott Park, IL
Objectives: Fast LC has recently gained a lot of popularity by reducing the LC/MS/MS cycle time during
bioanalysis. One risk associated with fast LC is the loss of resolving power under high flow rate. Two
columns designed to achieve fast separation without compromising resolution were used for 4
development compounds in 7 validations. The success and challenges of using both the superficially
porous column (also called as fused core silica column) and the small particle column (2.5 µm) in GLP
environment will be presented. Methods: All the validations were performed on an API 4000 mass
spectrometer. A Shimadzu SIL HT autosampler was used with a Shimadzu 10ADvp HPLC. The run
time was less than a minute with a total LC cycle time of less than ~1.5 minutes for all methods. All
validated methods were used to support GLP tox studies or clinical studies. ISR (Incurred Sample
Reproducibility) was performed for all the studies analyzed with fast LC-MS/MS methods. Results:
Both the superficially porous and the small particle column provided satisfactory validation results and
acceptable ISR results. However, the superficially porous column (Halo C18) is more robust compared to
the small particle column (Phenomenex Max-RP). An interfering conjugated metabolite peak was
observed during GLP tox sample analysis. A modification of mobile phase had to be made to resolve the
two peaks for accurate integration. Therefore, risks associated with running fast LC should be taken into
account when evaluating methods to be used for regulated studies.
Abstract 155
A uHPLC-MS/MS Method for the Quantitation of a Drug Candidate in Rat Plasma with Three
Drug-related Compounds Present
Heidi M. Snapp; Guowen Liu, Anne F. Aubry, Qin C. Ji, Bioanalytical Sciences, Research &
Development, Bristol-Myers Squibb Co. Princeton, NJ

Introduction: Ultra High Performance Liquid Chromatography (uHPLC) using sub-2 µm columns has
become a popular technique for bioanalytical assays using LC-MS/MS. One reason is that it can greatly
increase sample throughput but still maintain the required separation for a given analyte. Here we report a
fast uHPLC-MS/MS bioanalytical method for the quantitation of a drug candidate in rat plasma. Based on
the discovery data, three drug-related compounds may be present in incurred samples. Therefore, a good
separation of the analyte from these interference compounds is critical for the success of this method.
Methods: This method utilizes a Leap 4x Ultra (Flux) pump to deliver the mobile phases.
Chromatographic separation was achieved with gradient elution on a Waters Acquity UPLC BEH C8, 1.7
µm, 2.1 × 50 mm column. The mobile phase contained water, acetonitrile, ammonium bicarbonate and
ammonium hydroxide. Detection was accomplished using a Sciex API 4000 tandem mass spectrometer in
positive ion electrospray SRM mode. Results: In this method, four compounds were well separated with
a run time of 1.5 minutes. This assay showed excellent accuracy, precision and reproducibility. The intra-
assay and inter-assay precision was within 4.4% CV and 4.8% CV respectively. The assay accuracy was
within 3.3% of the nominal concentration values. In the incurred sample reanalysis test, the reanalysis
results were within 10% of the mean of the repeat and the initial values for all 20 samples tested. This
assay has been successfully applied to multiple GLP toxicology studies.

Abstract 156
Validation of a Liquid Chromatography Tandem Mass Spectrometry Method for the
Determination of Latanoprost Acid in Plasma From Pediatric Subjects

JoAnn La Fargue1, John Smeraglia1 Bruce Hidy2, Mike Modisett2; 1Pfizer, PGRD La Jolla Laboratories,
Pfizer Inc., La Jolla, CA, 2PPD Inc., Richmond VA.
Objective: To develop and validate a sensitive LC-MS/MS method (10 pg/mL) to determine latanoprost
acid (LA) concentrations in plasma from pediatric subjects. Latanoprost is the ester prodrug of
latanoprost acid used to treat ocular hypertension. Methods: Liquid-liquid extraction was applied to 0.5
mL of plasma fortified with latanoprost acid-d4 as internal standard (IS). The organic layer was applied
to a SPE column and LA was eluted with a solution of formic acid, hexane, and isopropanol 2:40:60
(v/v/v). Using column switching, HPLC was performed with an Aquasil 2.1 mm x 50 mm, 3 µm loading
column and a Restek 2.1 mm x 50 mm biphenyl, 5 µm elution column at a flow rate of 250 µl per minute
with a 12 minute gradient starting at 50:50 acetic acid (0.005%) / acetic acid (0005%) in methanol (v/v).
Upon elution of LA from the HPLC, detection and quantification was achieved with a Sciex API 5000
mass spectrometer using electro-spray ionization (ESI). Negative ions were monitored in MRM mode
with m/z transitions for LA at 389.3 to 146.9 and IS at 393.3 to 147.1. Results: The method was
validated over a range of 10 to 600 pg/mL with correlation coefficients of >0.99. Inter-batch accuracy
(%RE) and precision (%CV) for quality control samples ranged from -7.16% to -0.392% and ≤ 10.1% for
accuracy and precision, respectively. Implications: A sensitive and accurate method to determine
latanoprost acid in human plasma was developed, validated and successfully applied to the quantification
of LA in clinical samples from pediatric subjects following ocular administration of latanoprost.
Abstract 157
A Novel Technique for Monitoring the Matrix and Metabolites while Simultaneously Detecting
Ibuprofen by MRM

Geneen Baynham, Paul D Rainville, Jose Castro-Perez, Joanne Mather, Robert S. Plumb, Waters
Corporation, Milford, MA
Objective: The detection of ibuprofen while simultaneously monitoring the matrix for potential
metabolites or interferences. Methods: A bioanalytical assay was used for the analysis of ibuprofen in
urine. Urine was taken from 2 subjects 8 hours after dosing with 400mg of ibuprofen. Simple sample
preparation by centrifugation and dilution was employed prior to injecting the samples on to an
LC/MS/MS system. Separation was performed on a 2.1 x 50mm, 1.7µm C column. The analytes were
detected by negative electrospray ionization in multiple reaction monitoring mode for the selective
detection of ibuprofen. The MS conditions were also set to allow the simultaneous acquisition of full ms
scan data. Results: The full scans acquired for each volunteer demonstrated the variability of matrix
between the two subject's urine. Furthermore, the full scan was mined for parents of likely metabolites.
Peaks with m/z values which corresponded to common metabolites were identified. The ibuprofen was
detected using the transition 205->161. Implications: Use of novel technology to simultaneously scan
the matrix of a bioanalytical sample whilst detecting, an analyte of interest will allow the scientist to be
aware of any changes in matrix which might affect the ruggedness or accuracy of a bioanalytical assay
and also to monitor for any unexpected metabolites of the analyte or of over the counter medications.
Abstract 158
Evaluation of 96-well Phospholipid Depletion Plates for LC-MS/MS Bioanalysis.
Joseph Tweed, Leena Khullar, Ying Zhang, Michael Furlong, Pharmacokinetics Dynamics and
Metabolism Pfizer Global Research and Development, Groton, CT
Objectives: Determine the effectiveness of phospholipid depletion (PLD) plate technology and assess
their utility for routine use in regulated bioanalytical assays. Methods: Human plasma was precipitated
on a standard 96-well Strata Impact protein precipitation (PPT) plate, a 96-well Varian NDlipids plate or a
Supelco Hybrid SPE plate. In each plate, 100 uL of human plasma was added to 300 uL precipitation
solvent (acetonitrile). The sample eluent was collected, evaporated to dryness under N2 at 40°C and
reconstituted in 0.2 mL of appropriate reconstitution solvent depending on chromatography conditions.
Samples were injected into an API4000 tandem mass spectrometer using positive electrospray ionization
in MRM mode. Hydrophilic interaction chromatography (HILIC) was used to determine the percent
depletion of the most prevalent phospholipids found in protein precipitation extracts. Reversed-phase
chromatography was used to assess the impact PLD plates have on ionization and matrix effect
experiments routinely conducted during assay validations. Data was quantified by measuring peak area
ratios. Results: Nearly 100% of endogenous phospholipids can be removed from protein precipitation
extracts. Pfizer test compounds demonstrated that ionization and matrix effect data could change
depending upon the use phospholipid depletion technology. Data indicate PLD plates significantly
improve matrix effect and ionization effect data when using an analog IS for drug quantitation.
Implications: Currently available phospholipid depletion plates do offer a practical solution for removing
phospholipids from precipitated human plasma samples. Using depletion plates to assay samples for
routine regulated bioanalytical assays can lead to improved ionization and matrix effect data.
Abstract 159
Method Development Strategies For Ultra High-Throughput LC-MS/MS Analysis of Small Polar
Molecules Utilizing HILIC Mechanisms

Brian Rappold; Russell Grant; Patricia Holland, Labcorp, Burlington, NC
Introduction: HILIC has proved to be a robust and selective tool (from isobars and phospholipids) for
analysis of small polar molecules, enabling ultra-high-throughput-LC-MS/MS methodologies (>2000
samples/instrument/day). This paper describes strategies wherein functional groups, sorbent-bed selection,
and non-intuitive mobile phase chemistries can be utilized to provide selective, suppression-free mass
spectrometric detection of small polar analytes. Methods: Nornicotine (NN), Succinic Acid (SA), and
Orotic Acid (OA), (all <200 amu) containing carboxylic acids and/or amines, were spiked into DI water
and human plasma at 500 ng/mL. 300µL sample was precipitated with 1200µL acetonitrile; samples were
vortexed, centrifuged, and 10µL supernatant was injected. 52 mobile phase modifications on 8 HILIC-
style columns were screened; comparison of retention times for analytes and phospholipids were used to
elucidate developmental workflows for the selective shifting of phospholipids and isobaric interferences.
Results: Chromatographic selectivity based on modifications in acid, buffer, and protic solvent
concentrations was evaluated by monitoring the retention times of analytes, and phospholipid classes
(phosphatidyl/lysophosphatidyl cholines). Protic organic solvent elution on silica and HILIC sorbent
beds contributes to the early elution of phospholipids and analytes lacking amines (SA), whereas the
retention of aminated (NN/OA) molecules is relatively unchanged with protic solvents in the presence of
buffering materials. Selectively shifting isobaric interferences and phospholipids from carboxylic acids,
regardless of amine content, is achieved by utilizing ion exchange mechanisms derived from volatile
buffers, charge state modifications of phospholipids in solution and aqueous content in gradient
chromatography. Stepwise developmental workflows will be shown; enabling these mechanisms to be
elucidated.
Abstract 160
The Gyrolab: A New Frontier in Immunoassay?

Matthew Szapacs, GSK, Jonathan Kehler GSK,, Scott Hottenstein,GSK, William Lorelli, GSK, and
David Citerone, GlaxoSmithKline, King of Prussia, PA
Objectives: The use of enzyme-linked immunosorbant assay (ELISA) has been the gold standard for the
quantitation of biomolecules from complex media. In general an ELISA is carried out over a two day
period of time where multiple reagents are incubated for varying lengths of time in a 96-well plate.
ELISA methods are normally very selective and sensitive however the range is often limited to less than
two orders of magnitude making multiple dilutions mandatory for the analysis of toxicokinetic studies
where plasma concentrations can exceed 1 mg/mL. Methods: The Gyrolab is a compact disc based
system that performs all steps of an immunoassay for 112 samples in 50 minutes with detection by laser
induced fluorescence and allows ranges on the order of four orders of magnitude. Results: To test the
utility of the system, we converted an in-house ELISA method for the quantitation of a humanized
monoclonal antibody (range 100 to 2500 ng/mL) to the Gyrolab platform. The Gyrolab allowed a range
of 10 to 10,000 ng/mL in rat plasma (with a maximum %CV of 14.1% and a maximum Bias of -11%) and
250 to 100,000 ng/mL in human plasma (with a maximum %CV of 8.3 and a maximum Bias of 13.6%).
Implications: The Gyrolab is an easy to use fully automated way to run immunoassays. It allows the
user to speed up method development time, increase sample throughput and decrease the amount of time
spent diluting samples.

Source: http://ce.pharmacy.wisc.edu/courseinfo/archive/2009JulyLOL/Abstracts%20for%20the%20website%206-22.pdf