Transcript 2003 Colloquium Poster

Simultaneous UPLC-TMS assay for the detection of Levetiracetam and Gabapentin in serum and plasma.
K. Johnson-Davis1,2, J.M. Juenke1, P.I. Brown1, G.A. McMillin1,2
1ARUP
Institute for Clinical and Experimental Pathology, Salt Lake City, UT, 2Department of Pathology,
University of Utah School of Medicine, Salt Lake City, UT
Methods and Results
(KeppraTM)
(NeurontinTM)
Background: Levetiracetam
and gabapentin
are anticonvulsants with novel
structures, and suggested therapeutic ranges of 5 to 30 µg/mL and 2 to 10 µg/mL, respectively.
Immunoassays are not currently available to support TDM of these anticonvulsant drugs. Here we describe
a simple, rapid assay for the simultaneous monitoring of levetiracetam and gabapentin in plasma by UPLCMS/MS. Methods: After a protein crash with 50:50 methanol:acetonitrile internal standard solution, 1 µL of
supernatant sample is injected onto a Acquity UPLC HSS T3 1.8 um, 2.1 x 50 mm (Waters) column. Elution
occurs using a linear gradient of acetonitrile and water each having 0.1% formic acid added. The
compounds are eluted into a Waters Acquity UPLC TQD, operating in a positive mode to detect gabapentin
at transition 172.18>154.11, levetiracetam at 171.11>126, and internal standard (3-Amino-2-naphtholic acid)
at 188.06>170. Secondary transitions for each analyte are also detected for gabapentin at 172.18>137.06,
levetiracetam at 171.11>154, and internal standard at 188.06>115. Runtime is 1.5 minutes per injection.
Due to similarity in molecular weights between the compounds chromatographic separation was achieved.
Results obtained from patient samples were compared to results generated by established HPLC-UV
methods. Results: Method validation showed an intra-assay imprecision (CV) below 8% and an inter-assay
CV below 5% for both analytes at four different values across the range with total imprecision less than
14%. Comparison data with HPLC-UV was excellent, as measured by a linear regression of y = 1.12x –
0.77, r=0.996, Sy,x =0.89 for gabapentin, and y = 0.991x +0.70, r=0.997, Sy,x =2.24 for levetiracetam. The
analytical measurement ranges were 1 to 150 µg/mL for gabapentin and 5 to 150 µg/mL for leviteracetam,
and no analytical interferences were identified. Conclusion: A simple, reliable TMS method was developed
and validated for routine clinical monitoring of levetiracetam and gabapentin.
Background
Gabapentin:
Novel anticonvulsant drug for adjunctive therapy of partial seizures.
Approved for use in children.
Structurally similar to gamma-aminobutyric acid (GABA).
Protein bound fraction of about 3%, excreted via the kidneys
Elimination half-life of approximately 6 hours and clearance proportional to creatinine clearance
Used in pain management, alcohol withdrawal, migraine therapy
TQD Assay:
The standards, controls, and patients were precipitated using a 50:50 methanol: acetonitrile mixture
that contained the Internal standard (IS) 3-amino-2-naphthoic acid. After the addition of this solution
the samples were vortexed and centrifuged. The organic phase was then poured into an
autosampler vial and analyzed by HPLC-MS/MS.
The Waters Acquity UPLC TQD ulilizes Masslynx software. The instrument utilizes ESI interface,
multiple reaction monitoring (MRM), and positive ion mode. The resolution of both quadrapoles
was maintained at unit mass resolution with a peak width at half height of 0.7 amu. The data
analysis was performed using the Water Quanlynx software.
Instrument settings:
LC Seperation:
Column: Acquity UPLC HSS T3 1.8 um, 2.1 x 50 mm (Waters)
Mobile Phase: Bottle A: Barnstead Nanopure water or equivalent with formic
acid. (4L water : 4 mL formic acid)
Mobile Phase: Bottle B: Burdick and Jackson or Fisher Optima Acetonitrile
(ACN) with formic acid. (4L ACN : 4 mL formic acid)
Mobile phase gradient:
Time` Flow rate
%A(water)
%B(ACN)
0.00
0.65
95
5
0.10
0.65
95
5
0.80
0.65
60
40
0.85
0.65
10
90
1.00
0.65
10
90
1.05
0.65
95
5
1.50
0.65
95
5
LC column temperature: 24C (Room Temperature).
Mass Spec settings:
Parent (m/z)
Daughter (m/z) Dwell(sec) Cone (volts) CollisionEnergy(eV)
171.11
126.00
0.020
14
14
171.11
154.00
0.020
14
8
172.18
137.06
0.020
26
14
172.18
154.11
0.020
26
14
188.00
115.00
0.020
23
30
188.00
170.00
0.020
23
10
 A majority of patients at suggested doses fall within a 2-10 mg/L range chronic pain and addictions
higher trough concentrations of 15-30 mg/L are maintained. (1)
Levetiracetam:
Novel anticonvulsant drug for adjunctive therapy in partial-onset seizures in adults
Sensitivity:
Rapidly absorbed after oral administration, with peak plasma levels at 1-1.5 hours.
Table 1: Imprecision for gabapentin and levetiracetam (n=24)
Method sensitivity was evaluated by analyzing standards progressively lower in concentrations.
Limit of quantitation (LOQ) was determined as the lowest concentration at which accuracy was
within 20%, imprecision was within 15%, and the signal to noise ratio for the primary ion
transitions were greater than 5. In two separate studies LOQ was determined to be 1 g/mL and
the limit of detection (LOD) was 1 g/mL. Concentrations below this limit should be reported as a
less than 1 g/mL.
Method Correlation:
Human samples received for routine testing were used for the evaluation of the method accuracy
and de-identified according to corporate policy. The samples included in this study were analyzed
using the in house HPLC-UV assays and the TQD assay. Twenty-nine comparisons were done for
each assay. The regression equations, correlation coefficient and standard error for the
comparisons were for Gabapentin: ARUP TQD= 1.12 HPLC(UV)-0.77, r = 0.996, Sy,x=0.89. For
Levetiracetam: ARUP TQD= 0.991 HPLC(UV)+0.70, r = 0.997, Sy,x=2.24. The data indicated that
no significant bias existed between the methods.
50
45
40
35
30
25
20
15
10
5
0
Conclusions
•The instrument analysis by the method is based on unique fragmentation of the molecules
and eliminates potential interference observed in other methodologies.
90
80
Gabapentin
Y= 1.12x-0.77
r = 0.996
r2 = 0.992
Std. error = 0.89
0
Chemically unrelated to the existing anticonvulsant drugs.
The method imprecision was determined by analyzing a total of four controls each for each
analyte. The four controls were homebrew spiked samples in drug-free plasma. The
gabapentin homebrew controls were spiked at 2, 3.5, 8, and 16 g/mL respectively and the
Levetiracetam controls were spiked at 12, 20, 65, and 100 g/mL respectively. The
homebrew controls were independent for each drug. Gabapentin and Levetiracetam
concentration in the sample (n=24), within-run, between-run, and total imprecision for the
results obtained in the experiment are presented in Table 1.
The method linearity was evaluated by analyzing standards prepared at 1,10,30,50 and 150 g/mL
of gabapentin and levetiracetam Each standard was analyzed in duplicate and concentrations were
calculated from the standard curve using concentrations of 1, 10, 30, 50, and 150 g/mL. The
standards and spiked samples were made from separate stock standards. The assay was found to
be linear up to 150 g/mL, with an inaccuracy better than 0.45% at the highest evaluated
concentration of 150 g/mL for gabapentin and linear up to 150 g/mL, with an inaccuracy better
than 0.25% at the highest evaluated concentration of 150 g/mL for levetiracetam.
TQD
Plasma gabapentin concentrations and for clinical effect within the dosage range 600 to 1800 mg/day.
Recovery and Imprecision:
Linearity:
TQD
Abstract
10
20
30
HPLC
40
50
60
•Advantages to this method include simple sample preparation, and fast instrument analysis
(1.5 minutes per sample).
70
60
50
40
30
20
10
0
-10 0
References
Levetiracetam
Y= 0.991x+-0.7
r = 0.997
r2 = 0.996
Std. error = 2.24
20
40
60
80
100
HPLC
1. Juenke J.M., Brown P.I., McMillin G.A.and Urry F.M. Procedure for the monitoring
of gabapentin with 2,4,6-trinitrobenzene sulfonic acid derivatization followed by HPLC
with ultraviolet detection. Clin Chem, 2003 49(7) 1198-1201.
2. Juenke JM, Brown PI, Urry FM, McMillin GA. A procedure for the monitoring of
levetiracetam and zonisamide by HPLC-UV. J Anal Toxicol 2006 Jan/Feb 30(1):27-30.
Not extensively metabolized - produces only one non-reactive carboxylic metabolite.
Figure 2: Comparison plots for Gabapentin and Levetiracetam in patient samples. All results are in g/mL.
No drug-drug interactions have been described.
Half-life of 6-8 hours
INTEREFERENCE:
Primarily eliminated in urine; therefore, dosing may be adjusted based on renal function.
Tentative target therapeutic range is 6.0 to 20 µg/mL.
Structurally related compounds were evaluated for potential interference as were coadministered
drugs. These included other anticonvulsants such as primidone, phenobarbital, lamotrigine,
oxcarbazine and metabolite, ethotoin, methsuximide, mephenytoin, zonisamide, felbamate, and
topiramate. Other drugs tested included the tricyclic antidepressants and several benzodiazepines, as
well as caffeine, ibuprofen, acetaminophen, and other over the counter drugs. Several drugs of abuse
were also tested including amphetamine, methamphetamine, methadone and cocaine. The drugs
were tested at the stock concentrations of 1 g/L. None were found to interfere.
Under clinical study for refractory epilepsy, benzodiazepine withdrawal, monoclonus,
and mood stabilizing properties. (2)
Figure 1: Chromatograph of Assay MRMs.
Acknowledgment
We acknowledge the assistance of the staff in the ARUP Clinical Toxicology Department
and the ARUP Institute for Clinical and Experimental Pathology