Development, optimization and validation of a HPLC DAD method assaying Levetiracetam and common AED in serum

(1)

Final report

Sdffdfsdf

Development, optimization and validation of a HPLC DAD

method assaying Levetiracetam and common AED in serum.

Sander de Koning

(2)

Version: 1

Edited last: August 7, 2013

Group: Avans University of Applied Sciences Bachelor of Science: Chemistry

Major: Forensic Laboratory Science

Author: S. de Koning (s.dekoning@student.avans.nl)

(sanderd.koning@gmail.com) Supervisors: J.M.C Konig-Quartel (j.m.c.Konig-Quartel@lumc.nl)

Dr. J. den Hartigh (J.den_Hartigh@lumc.nl) Dr. BM de Rooij (bm.derooij@avans.nl)

Location: Leids Universitair Medisch Centrum

Albinusdreef 2 2333 ZA Leiden The Netherlands

Educational institution: Avans University of Applied sciences

Academie voor Technologie, Gezondheid en Milieu

Lovensdijkstraat 61‐63

4818 AJ Breda The Netherlands

Final report

Sdffdfsdf

Development, optimization and validation of a HPLC DAD

method assaying Levetiracetam and common AED in serum.

Sander de Koning

(3)

Acknowledgements

Before you lies the final report written during the end and after my graduation internship period at the Department of Clinical Toxicology and Pharmacy, at Leiden University Medical Centre (LUMC).

My interest in the pharmacies laboratory was raised during my previous internship at Erasmus University Medical Centre at the Department of Clinical Chemistry. During this current internship my interest for toxicology has been further developed and I am even considering and currently looking in to continue my studies in Biomedical Sciences or, even more so, in Pharmacy.

For the establishment of this final report I would like to thank my mentor Jacqueline Konig for putting up with me for the past half year and passing on her detailed knowledge of chemistry and instruments used during this internship. I would also like to thank my family and friends for letting me work in peace during the nightly hours and supporting me all the way.

Finally I would like to enounce my gratitude for everyone at the department and colleague interns that have assisted me in any way during my stay with you.

- Now it's your turn to graduate, mum! - Sander de Koning

(4)

1 Abstract

Levetiracetam is a relatively new anti epileptic drug. Although manufacturers state that therapeutic drug monitoring is not necessary The Department of Clinical Toxicology and Pharmacy at Leiden University Medical Centre is encountering a sustained amount of increase of analysis requests of patient samples with Levetiracetam. As there was no in-house Levetiracetam assay, the analysis requests were all outsourced to an external laboratory.

This report describes a gradient high-perfomance liquid chromatographic (HPLC) method for the simultaneous measurement of Levetiracetam, Oxcarbazepine (OXCZBZ), 10-hydroxy dihydrocarbamazepine (10-OH-CBZ), Carbamazepine (CBZ), Carbamazepine-10,11-epoxide (CBZ-Epoxide), Dihydroxy dihydrocarbamazepine (CBZ-Diol), Phenytoin, Phenobarbital and Lamotrigine in human serum (200 µl) utilizing β-Hydroxy ethyl theophylline and Hexobarbital as internal standards. Monitoring these drugs is important, e.g. for detecting potentially toxic concentrations and therapeutic drug monitoring.

Following a liquid-liquid extraction, Levetiracetam (5-50 mg/L, R2_{=0.9999), Phenobarbital}

(5-50 mg/L, R2_{=0.9997), Lamotrigine (2-20 mg/L, R}2_{=0.9982), CBZ-Diol (1-10 mg/L,}

R2_{=0.9995), 10-OH-CBZ (5-50 mg/L, R}2_{=0.9996), CBZ-Epoxide (0,5-5 mg/L, R}2_=0.9998),

OXCBZ (1-10 mg/L, R2_{=0.9997), Phenytoin (3-30 mg/L, R}2_{=0.9997) and CBZ (2-20 mg/L,}

R2_{=0.9999) were seperated on a C18 ODS Hypersil column and eluent was monitored by}

Diode Array Detection at a wavelength of 210 nm. This multi anti epileptic drug assay showed inter- and intra-assay precision and accuracy values within ±15%. Matrix effects on different batches of serum, displayed accuracy within the required 10% correlation of variance (CV). Although the method showed unexpected recovery yields of 67-133%, the yields were stable. All analytes were stable at room temperature, +4˚C and -20˚C for at least 15 days; except for OXCBZ that must be analyzed straightaway or be kept at -20˚C until analysis. LLOQ (within ±20% accuracy and precision) was determined at 1.25 mg/L for Levetiracetam and Phenobarbital, 0.67 mg/L for Lamotrigine, 0.20 mg/L for CBZ-Diol, 1 mg/L for 10-OH-CBZ, 0.05 mg/L for CBZ-Epoxide, 0.10 mg/L for OXCBZ, 0.75 mg/L for Phenytoin and 0.40 mg/L for CBZ. All analytes were linear (within ±10% precision) at at least 300% of the highest calibration standard, with the exception of CBZ-Diol limited to 10 mg/L.

In conclusion this method is suitable for routine analysis of the previously mentioned anti epileptic drugs and their metabolites.

(7)

2 Introduction

This chapter contains and explains the necessary theory to accompany and complement this project and its objective. The first part will provide general information about the department where the project was executed. The disorder that is treated with Levetiracetam is then covered followed by Levetiracetam itself and other drugs of interest. Subsequently the analytical techniques utilized to accomplish the aim of the project.

2.1 Pharmacy

An academic hospital is a hospital that, other than providing general medical care to patients, has an emphasis on providing training and education to future and current health professionals. Although surgeons, cardiologists and nurses may be the first to come to mind, there are numerous other specialists with their own departments within a hospital. The hospital’s pharmacy is one of those departments.

A pharmacy is not necessarily a desk where assistant pharmacists provide clients with (un)prescribed medicine. A pharmacy also provides guidance and information, checks whether other prescribed drugs could react. An academic hospital usually has a pharmacy department that does more than just provide the prior.

2.1.1 Academic hospital pharmacy

Academic hospital pharmacies are usually equipped and accredited to prepare medicine from raw materials. These drugs are then used to treat patients. To ensure the purity and quality of the prepared drugs, the pharmacy is also equipped with a quality control lab that subjects the drugs and raw materials to quality control analysis. Pharmacogenetics also plays an important role in the pharmacy. It is the science of differences in metabolic pathways that are expressed in polymorphisms in genes. These differences can result in a different, positive but also adverse, response to drugs. When clinical toxicology values are co-related to certain polymorphisms (or single nucleotide polymorphisms), a prediction can be made as to what the (adverse) effects will be at certain dosages, prior to treatment. This results in more accurate medication which subsequently reduces the cost of medical care.

(8)

2.1.2 Clinical toxicology

Additionally, and in the case of this project, clinical toxicology is performed within the academic hospital pharmacy. Within the section of clinical toxicology, analysis is performed to determine presence and or levels of drugs in body fluid. To be precise, unlike clinical chemistry which determines endogenous substances, clinical toxicology determines non endogenous substances. Clinical toxicology is performed for therapeutic drug monitoring, analysis of (unknown) intoxications. Numerous analysis methods are used to achieve the quanti- and qualification of these non-endogenous substances. Immunoassays, gas- and liquid chromatography with several kinds of detectors ranging from diode array detectors to mass spectrometers are at the disposal of the analysts to analyze the biological samples. [1]

2.2 Epilepsy

Epilepsy is a common (prevalence rate of 1 in 200) neural disorder of the brain in which an individual has seizures that repeat over time. Therefore an individual with a single seizure (also acute seizure) is not considered to have epilepsy. These seizures are also called attacks, fits and convulsions. During these convulsions involuntary electrical discharges cause changes in the body’s actions. These electrical discharges can be observed by electroencephalography Figure 2.1 [2-5]

Figure 2.1, typical electroencephalogram of a patient with Idiopathic epilepsy.

(9)

The cause of epilepsy is in most cases unknown. Though there are many possible causes known, which include but are not limited to, abnormal brain development, illness, brain injury and even genetic. Although the abnormalities in genes do not directly lead to epilepsy, it does increase the possibility of the brain being subject to seizures. Chemical imbalances can also cause seizures, e.g. use of alcohol, cocaine, but also deficiencies in blood sugar and blood calcium levels. There are several types of seizures that can also be divided in to two other groups, partial and general epilepsy. In the following paragraph I will discuss the two major known types of epilepsy. [6]

2.2.1 Idiopathic epilepsy

With idiopathic (a medical term for unknown) epilepsy the nervous system appears to be normal, no abnormalities can be identified with the standard neural observation studies. This type of epilepsy is commonly treated with medication and some people even outgrow this disorder with the result of the ceasing of seizures. [7]

2.2.2 Symptomatic epilepsy

Symptomatic epilepsy arises from the effects of an epileptic lesion. This so called lesion is the result of an area of the brain acting abnormal. A lesion can have several causes such as a tumor, a lack of oxygen traveling to the brain, an infection, fetal development issues or a defect in metabolism causing widespread injury to the brain. This abnormality allows synchronous electrical discharges in the brain that eventual induce seizures. Symptomatic epilepsy can be treated with medication, to eliminate or suppress the seizures. If medication has little to no (desired) result, surgery may also be an option to eliminate the occurrence of seizures. [8, 9]

2.3 Antiepileptic drugs

As described in the epilepsy chapter, many situations can induce a seizure. Antiepileptic drugs can be grouped by their mechanism of action. Sodium channel blockers and calcium channel blockers are two examples of these groups. [1, 10-12]

2.3.1 Calcium channel blockers

A neural calcium channel is known to exist in three forms, namely L, N and T. Calcium channels work as a regulator of rhythmic brain activity. A certain seizure called absence seizure has been correlated to the so called “3 per second spike and wave discharges” that T-calcium channels have been known to play a role in. Particular antiepileptic drugs can prevent these T-calcium channels from producing excessive discharges. [10-12]

(10)

2.3.2 Sodium channel blockers

A sodium channel has 3 states: active, resting and inactive. When there is the potential of an action, for example movement, these channels are in an active state and allow passage of sodium ions (Na+_{). This passing of ions triggers nerves. Once the activation is}

terminated, or the potential of an action has passed, an amount of these channels become inactive. This is also called the refractory period. During a seizure sodium ions get fired rapidly constantly stimulating these channels. Certain antiepileptic drugs have the effect of stabilizing these channels to the inactive state; this assures the nerves are not rapidly triggered without a conscious cause. [10-12]

2.4 Levetiracetam

Levetiracetam is a relatively new anti-epileptic drug. It is derived from piracetam which is widely used for its assumed advantageous effects on cognition in the elderly and is the S-enantiomer of etiracetam also a nootropic drug.

Levetiracetams chemical structure (Figure 2.2) differs from other antiepileptic drugs. Gower et al. (1993) reported that by routine (random) research on rodents it came to light that Levetiracetam had a potent anticonvulsant effect against audiogenic seizures, electrically induced convulsions and convulsions induced chemically. Not long after these findings were confirmed in other papers (e.g. Löscher et al. 1993), UCB Pharmaceuticals started clinical trials on epilepsy patients. These trials resulted in the United States Food and Drug Administration (FDA) approval in December 1999 and the European Medicines Evaluation Agency in September 2000. [12-17]

Figure 2.2, structure of Levetiracetam.

(11)

2.4.1 Mechanism of action

Although the mechanism of action of Levetiracetam is still not fully understood, it is however known that it does not (solely) work by the way that classic antiepileptic drugs work. R. Surges et al. (2008) reviewed its mechanism of action in epilepsy and summarized the findings of numerous (recent) papers on voltage-gated ion channels, regulation of intracellular ions and synaptic transmission. The review concluded that although it does seem to share some targets with other antiepileptic drugs, it also shows new and unparalleled binding sites but also mechanisms of action that have not yet been reported with prior classical antiepileptic drugs. It is used and is effective in not only generalized but also partial epilepsy syndromes. Levetiracetam is not only used for monotherapy but can also be prescribed as add-on medication. [12, 18]

2.4.2 Pharmacokinetics and metabolism

Levetiracetam is completely (>95%) and rapidly (peak concentrations within 1 hour) absorbed after its oral administration. It shows linear (and thus

highly predictable) pharmacokinetics and endures minimal (<10%) plasma protein binding. The elimination half-life of Levetiracetam is age dependent. In children 6-12 years old it takes 6 hours. In adults it is longer, approximately 8 hours. Levetiracetam is minimally metabolized; it has one major metabolite (Figure 2.3). Its metabolites are all inactive, which makes Levetiracetam favorable for add-on therapy. Metabolism is performed by the common cytochrome P450, but is enzymatically hydrolyzed within cells. [19-21]

Figure 2.3, structure of 2-pyrrolidone N-butyric acid, the major (inactive) metabolite of Levetiracetam.

(12)

2.5 Other antiepileptic drugs and metabolites of interest

Because it was preferred to create a multidrug assay, table 1 shows the other antiepileptic drugs and their (active) metabolites that will be quantified with this assay.

Table 1, other antiepileptic drugs and metabolites of interest

Name Abbrev. used Chemical

formula LogP --- Mw Chemical structure Oxcarbazepine OXCBZ C15H12N2O2 1.7 --- 252.27 10-hydroxy dihydrocarbamazepine 10-OH-CBZ C15H14N2O2 1.4 --- 254.28 Carbamazepine CBZ C15H12N2O 2.5 --- 236.27

(13)

Carbamazepine-10,11-epoxide CBZ-Epoxide C15H12N2O2 1.3 --- 252.27 Dihydroxy dihydrocarbamazepine CBZ-Diol C15H14N2O3 0.3 --- 270.28 Phenytoin - C15H12N2O2 2.5 --- 252.27 Phenobarbital - C12H12N2O3 1.5 --- 232.24 Lamotrigine - C9H7Cl2N5 1.4 --- 256.09

(14)

2.6 Analytical techniques

It is possible to analyze Levetiracetam in blood by use of high performance liquid chromatography (HPLC) and gas chromatography. However, HPLC is chosen due to the availability of equipment and that a current HPLC antiepileptic drug assay is already developed and validated. To ensure quick and efficient sample preparation, liquid-liquid extraction (LLE) is performed. The following paragraphs will provide basic information for the techniques to be used and will be related to the analysis of Levetiracetam.

2.6.1 Liquid/liquid extraction

Because it isn’t possible to simply inject an amount of serum into the HPLC system and quantify Levetiracetam, the serum (or sample) needs to be pretreated. The serum sample is transferred together with an acidic buffer (the aqueous phase); in this case a phosphate buffer with pH of 4.0, in to a test-tube. The organic phase is dichloromethane. The target analytes move to the organic phase after vortexing the test-tube. The following step is to separate the layers. This separation is based on the density of the different fases. It is achieved by centrifuging the test-tube. After removing the top aqueous layer, the organic phase is transferred to a new test-tube where it is subjected to vaporization under a gentle stream of nitrogen. The dry residue is then redissolved, now with a solvent that is the same as the mobile phase of the HPLC system.

2.6.2 High performance liquid chromatography

HPLC is a chromatographic separation technique that provides the ability to separate, identify and quantify compounds of a mixture. HPLC is important because most targets are not volatile (enough) for gas chromatography, as is Levetiracetam with a boiling point of 396°C. In this case it is used to quantify Levetiracetam in the solution after LLE extraction. [1, 22]

As demonstrated in Figure 2.4 the HPLC system consists of a pump with degasser to supply the mobile phase under pressure coming from the three used eluent reservoirs. The next step in the HPLC is the injector. This setup has an auto analyzer. The benefits of the use of an auto analyzer are that the way of injecting the sample is not subjected to human error or variabilities with injecting samples and it is possible to continue other activities after setting up a batch because one does not need to manually inject every sample. One of the most important components of the HPLC system is the column. [1, 22]

(15)

The column is responsible for the separation of the compounds in the sample. A C18 column is chosen for its wide applicability in target analytes. As per reference F. lancelin et al. (2008) reported a HPLC method for the detection of Levetiraceteam with a C18 column. The column used in this article is however longer (250mm). The choice to start with a shorter column is due to the fact that a longer column would delay other target antiepileptic drugs too much, making the run time too long. In front of the column a pre column (also C18) is installed to protect the analytical column. [1, 22]

The detector used for the analysis of Levetiracetam is a Diode Array Detector (DAD). It is possible to detect five specific wavelengths with the DAD and an additional 3D UV spectrum. The principle of the DAD is based on absorbance of UV light. The more a molecule absorbs the light at that wavelength, the higher the output of the detector. The mobile phase should not be absorbent at the wavelength that is being detected. Gas in the mobile phase also interferes with the detection; therefore a degasser is used to remove this interference. [1, 22]

(16)

2.7 Method validation

To ensure the quality of an analytical method it is required to be validated. The validation is required to ensure the fact that the method is suitable and reliable for its intended use in routine clinical toxicological analysis and specifically therapeutic drug monitoring. The results from a validation process are then used to assess the quality, reliability and consistency. It is an important part of the Good Laboratory Practice standards. [22-24] The influence of the matrix (in this case serum) is evaluated by extracting and subsequently analyzing a low and a high standard in at least 6 different matrixes. The averages and coefficient of variances (CV) of detector responses are calculated. The CV is a measurement for the influence of matrices in relation to the extraction. [23, 24]

The recovery is determined in the lower and higher concentration area of the method. Therefor 6 low and 6 high standards are prepared. The response is compared to the response of 6 standards that are spiked with the exact same concentration would have been added prior to the extraction. This comparison represents recovery (or loss) of analyte after extraction in percentage. [23, 24]

The Precision is determined by comparing the exact amount (weight) of analyte added to matrix with the detected amount (response) that is calculated with the calibration curve in percentage. [23, 24]

The precision describes the scatter of results of repeated measurements of a homogenous sample. The intraday precision describes the degree of repeatability of the analytical method within the same ‘run’. The interday precision describes the degree of repeatability of the analytical method over several days. This shows the repeatability of the analytical method over a short and longer period of time. [23, 24]

The stability validation parameter shows the degree of stability of a sample in different methods of storage. Ambient temperature, refrigerated and frozen storage is analyzed. These values are compared over increasing periods of storage and will eventually describe the preferred method of storage for a specific time, or whether samples need to be analyzed straight away. [23, 24]

Linearity is determined by analyzing concentrations to max 150% above the highest standard. This upper limit of quantification is determined if the following higher concentration deviates more than 10% and/or shows a precision higher than 15%. [23, 24]

(17)

The lower limit of quantification represents the limit or confidence interval (95%) when a method can reasonably tell the difference between noise and an actual component. [23, 24] The selectivity of a method is determined by the ability to differentiate an analyte from a possible interfering component, e.g. metabolite, impurity and matrix component, which is expected to be present in the sample. [23, 24]

The stability of a method is determined by changing certain variables in a method with the purpose of checking whether these changes influence the methods quantitative results. [23, 24]

2.8 Overall objective

Levetiracetam has been approved by the US Food and Drug administration and the European Medicines Evaluation Agency just over 10 years ago. It is therefore a relatively new drug. Although the manufacturer states that monitoring of Levetiracetam is not necessary, the clinical toxicology lab of the Leiden Medical University Centre is observing an increase in amount of requests for analysis of Levetiracetam in blood. At the moment Levetiracetam analysis requests are outsourced to an external laboratory. It is therefore desirable that an in-house Levetiracetam assay is introduced.

(18)

3 Materials and Methods

3.1 Chemicals and Reagents

Levetiracetam (UCB Pharma, Breda, The Netherlands), Phenobarbital (LUMC, Leiden, The Netherlands), Lamotrigine (GSK Pharmaceuticals, Zeist, The Netherlands), CBZ-Diol (Novartis Pharma, Arnhem, The Netherlands), 10-OH-CBZ (Novartis Pharma, Arnhem, The Netherlands), CBZ-Epoxide (Novartis Pharma, Arnhem, The Netherlands), OXCBZ (Novartis Pharma, Arnhem, The Netherlands), Phenytoin(LUMC, Leiden, The Netherlands), CBZ (Sigma Aldrich, Zwijndrecht, The Netherlands), HET (Sigma Aldrich, Zwijndrecht, The Netherlands), and HEX (LUMC, Leiden, The Netherlands) were of analytical grade. Sodium Hydroxide, Phosphoric acid and Monopotassium phosphate were purchased from Merck (Amsterdam, The Netherlands) and were intended for analysis. Solvents used were all of HPLC grade. Methanol, Dichloromethane and Acetonitrile were purchased from Merck (Amsterdam, The Netherlands). Milli-Q demineralized water obtained from Millipore purification system from Merck Millipore (Amsterdam, The Netherlands).

Blank serum was obtained from Sanquin Amsterdam, Patient samples receiving AED treatment were obtained from routine analysis at The Department of Clinical Toxicology and Pharmacy at LUMC and kept at -20 ˚C. Stock solutions of Levetiracetam, Phenobarbital and 10-OH-CBZ were prepared at a concentration of 5 mg/mL in HPLC grade Methanol. Stock solutions of Phenytoin (3 mg/mL), Lamotrigine (2 mg/mL), Diol (1 mg/mL), CBZ-Epoxide (1 mg/mL), OXCBZ (1 mg/mL), CBZ (1 mg/mL), β-Hydroxy ethyl theophylline (HET) (1 mg/mL) and HEX (1 mg/mL) were also prepared in HPLC grade methanol. Stock solutions were stored at -20 ˚C, and were preservable for at least 1 year. The work solution (WS) and IS solution was prepared by adding the amounts described in table 2 to a 10 mL roundbottomed tube with a teflon closed cap (Sovirel, France), stored at -20 ˚C and were preservable for at least 1 year.

(19)

Table 2, Amounts of stock solution added for Work- and IS solution.

Work solution Volume_mL [Stock solution]_mg/mL IS solution Volume_mL [Stock solution]_mg/mL

Levetiracetam 1,00 5,00 HET 0,50 1,00 Phenobarbital 1,00 5,00 Hexobarbital 2,00 1,00 Lamotrigine 1,00 2,00 MeOH 7,50 - CBZ-diol 1,00 1,00 10-OH-CBZ 1,00 5,00 CBZ-epoxide 0,50 1,00 OXCBZ 1,00 1,00 Phenytoin 1,00 3,00 CBZ 2,00 1,00 MeOH 0,50 -

3.2 Calibration Standards and Quality Control Samples

Calibration standards were prepared by adding 2, 5, 10, 15, 20 µL work solution to 200 µL blank serum (Sanquin, Amsterdam, The Netherlands) resulting in the concentrations shown in table 3. Low and high quality control (QC) monsters were prepared by adding 2 µL and 20 µl WS to 200 µL blank serum. External QC's from Stichting KKGT (The Hague, Netherlands) were dissolved as instructed on packaging and 200 µL was used per QC sample.

Table 3, - concentrations of calibration curves in assay

Analyte Concentration in mg/L Levetiracetam 5 - 12,5 - 25 - 37,5 - 50 Phenobarbital 5 - 12,5 - 25 - 37,5 - 50 Lamotrgine 2 - 5 - 10 - 15 - 20 CBZ-diol 1 - 2,5 - 5 - 7,5 - 10 10-OH-CBZ 5 - 12,5 - 25 - 37,5 - 50 CBZ-epoxide 0,5 - 1,25 - 2 - 3,75 - 5 OXCBZ 1 - 2,5 - 5 - 7,5 - 10 Phenytoin 3 - 7,5 - 15 - 22,5 - 30 CBZ 2 - 5 - 10 - 15 - 20

3.3 Patient serum samples

Serum samples were collected from patients receiving treatment at the Leiden University Medical Centre as part of the routine clinical analysis and therapeutic drug monitoring and stored at 4 ˚C for the maximum duration of a month.

(20)

3.4 Sample preparation

To extract the AED's and IS's from serum, liquid-liquid extraction was conducted. Prior to this extraction 10 µL IS solution and 20 µL phosphatebuffer (pH 4,0) was added to calibration standards, QC's and patient samples and vortexed for 5 seconds. Subsequently 3 mL dichloromethane was added to each sample with the use of a positive displacement pipet. The sample was then vigorously vortexed for 30 seconds after which it was centrifuged for 4 minutes at 4000 rpm at room temperature. The supernatant and precipitated proteins were removed under vacuum after which the organic layer, containing the analytes of interest, was transferred to a clean pointy glass tube. Subsequently the newly transferred organic phase was then evaporated to dryness at 40 ˚C under a gentle stream of nitrogen. The dry residue is then subsequently reconstituted in 100 µl mobile phase, vortexed for 5 seconds and transferred to an auto sampler vial with insert.

3.5 Instrumentation and Chromatographic Conditions.

Chromatographic separations were achieved by using a Ultimate 3000 Quaterny Analytical pump, Ultimate 3000 Diode Array Detector and a Gina 50 auto sampler (Dionex - Thermo Scientific, Breda, The Netherlands). Chromatography was performed on a C18 ODS Hypersil 3µ column (50 x 4,6 mm) equipped with a column guard purchased at Thermo Scientific (Breda, The Netherlands). Mobile phase A consisted of 500 mL Methanol; B consisted out of 900 mL 0,1 M phosphate buffer (pH 6.8) and 100 mL Methanol. The flow rate was set to 0.8 ml/min with a gradient program displayed in table 4. 30 µL sample was injected in to the HPLC system and the eluent was monitored by diode array detection at 210 nm. 3D spectrum was also acquired from 200 to 340 nm for quality control. The HPLC system was controlled by Dionex Chromeleon 6,8 software (Thermo Scientific, Breda The Netherlands). Table 4, step gradient used.

Time (min) A B -1.0 0 100 1 0 100 16 50 50 18 50 50 20 0 100

(21)

3.6 Method validation

The method validation is derived from the internal method validation standards at the Department of Clinical Toxicology and Pharmacy at LUMC and does not differ from the procedures published by the FDA in "Guideline for submitting samples and analytical data for methods validation". [23]

All AED, metabolites and IS's were individually injected in to the HPLC system at a concentration of 10 mg/L to determine retention times.

Matrix effect was evaluated by analyzing low and high standards in six different batches of drug free serum.

Recovery was determined in the lower and higher concentration area of the method compared to the response of low and high QC monsters that were spiked with the exact same concentration in mobile phase.

Precision was determined by comparing the exact amount (weight) of analyte added to drug free serum with the detected amount (response), calculated with the calibration curve in percentage.

The inter- and intraday (n=6) precision were calculated by performing replicate analyses of QC's against a calibration curve.

Linearity is determined by analyzing standards in duplo at twice and four times higher concentration than the highest standard. [23, 24]

The lower limit of quantification was calculated by analyzing homogenous samples intraday (n=6) at concentrations ranging between 1/10 and 1/2 of lowest standard.

(22)

4 Results

4.1 Chromatography

Retention times of analytes injected individually are stated in table 6. Figure 5 shows a typical chromatogram of the lowest calibration standard.

Table 6, retention times of analytes

Analyte Time (min)

Levetiracetam 4.5

β-hydroxy-ethyl Thoephiline (IS) 5.8

Phenobarbital 10.4 Lamotrigine 10.9 CBZ-Diol 11.9 10-OH-CBZ 12.5 CBZ-Epoxide 12.8 Oxcarbamazepine 13.6 Hexobarbital (IS) 14.2 Phenytoin 14.9 Carbamazepine 15.5 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -50 -20 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320

35003072013 loq pat valid rec etc #2 [modif ied by sdekoning] standaard 1 UV _V IS_1

mA U min 1 - L evet iraceta m - 4,4 67 2 - 4 ,797 3 - b -hydrox y-ethy l theop hilline (IS) - 5,83 0 4 - F enobarbital - 10,3 77 5 - L amo trigine - 10 ,850 6 - C BZ-Diol - 11, 894 7 - 1 0-OH-C BZ 12,500 8 - C BZ-ep oxide - 12,7 54 9 - O xcar bazepi ne - 13 ,604 10 - Hexob arbit al (IS) 14,150 11 - Fen ytoïne - 14,8 87 12 - Carba maze pine - 15,547 WV L:210 nm

Figure 5, A typical chromatogram of the lowest calibration standard, a 200 µl drug-free

(23)

4.2 Quantification range

Linearity was determined by analyzing 5 calibration standards and a blank and is displayed as the following function "Y=AX+B" where Y is the concentration in mg/L, A is the slope and B is the offset. Table 7, linearity n=5 Analyte Function R2 Levetiracetam y = 6.8318 x + 0.1628 > 0.9999 Phenobarbital y = 18.6528 x + 6.8778 0.9997 Lamotrgine y = 20.8650 x - 8.3352 0.9982 CBZ-diol y = 19.3148 x - 1.4610 0.9995 10-OH-CBZ y = 29.5394 x + 14.9406 0.9996 CBZ-epoxide y = 45.9238 x + 0.1250 0.9998 OXCBZ y = 22.1220 x - 1.3294 0.9997 Phenytoin y = 23.3844 x + 4.8560 0.9997 CBZ y = 36.1898 x + 1.1766 >0.9999

Highest measurable concentration within 10% precision of the theoretical value was also determined and can be found in table 8. For reference the highest calibration standard in mg/L is noted per analyte.

Table 8, linearity above calibration curve n=2

Analyte Linear mg/L Highest calibration

standard in mg/L Levetiracetam At least to 150 50 Phenobarbital At least to 150 50 Lamotrgine At least to 60 20 CBZ-diol To 10 10 10-OH-CBZ At least to 150 50 CBZ-epoxide At least to 15 5 OXCBZ At least to 30 10 Phenytoin At least to 90 30 CBZ At least to 60 20

(24)

Lower limit of quantification can be found in table 9, CV and accuracy of the theoretical should be within 20% to pass the test at the given concentration.

Figure 6, A typical chromatogram of an LLOQ standard at 1/4th_{the concentration of the}

lowest calibration standard. The LLOQ of Levetiracetam, Phenobarbital and Phenytoin was determined at this level.

Table 9, linearity below calibration curve n=6

Analyte LLOQ (mg/L) _CV(%) _Accuracy(%)

Levetiracetam 1.25 19.2 112.7 Phenobarbital 1.25 20.0 97.9 Lamotrgine 0.67 3.3 115.2 CBZ-diol 0.20 0.2 116.3 10-OH-CBZ 1.00 14.0 88.1 CBZ-epoxide 0.05 19.7 85.3 OXCBZ 0.10 2.2 101.6 Phenytoin 0.75 15.3 81.6 CBZ 0.40 19.1 87.0 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110

12010072013 LOQ2 + HBH DA G LA TEN STA A N #10 [modif ied by sdekoning] standaard 1/10 UV _V IS_1 mA U min 1 - L evet iraceta m - 4,6 97 2 - 5 ,030 3 - b -hydrox y-ethy l theop hilline (IS) - 6,12 3 4 - F enobarbital - 10,9 87 5 - L amo trigine - 11 ,553 6 - C BZ-Diol - 12, 693 7 - 1 0-OH-C BZ 13,347 8 - C BZ-ep oxid e - 13,6 47 9 - O xcar bazepi ne - 14 ,573 10 Hexob arbit al (IS) 15, 117 11 - Fen ytoïne - 15,9 90 12 - Carba maze pine - 16,713 13 17,780 WV L:210 nm standaard 1/4

(25)

4.3 Matrix effect

The criteria for a suitable matrix is CV < 10%. As visible in table 10, all analytes at the lowest and highest calibration standard have passed this test.

Table 10, Matrix effect

Analyte Levetiracetam Phenobarbital Lamotrigine

Conc. (mg/L) 5.0 50.0 5.0 50.0 2.0 20.0 mean 0.2798 3.0482 0.8948 7.8447 0.4638 4.6390 SD 0.0063 0.1684 0.0070 0.2116 0.0056 0.1919 CV (%) 2.2 5.5 0.8 2.7 1.2 4.1 n 6 6 6 6 6 6 Suspect 0.268 3.365 0.906 8.162 0.456 4.938 Critical G 1.887 1.887 1.887 1.887 1.887 1.887 Calculated G 1.886 1.882 1.602 1.500 1.396 1.558 Outlier? No No No No No No

Analyte CBZ-Diol 10-OH-CBZ CBZ-Epoxide

Analyte OXCBZ Phenytoin CBZ

(26)

4.4 Stability

The criteria to define how long and at what temperature a sample should be stored, it is must comply to the following prerequisites: "CV < 15%" and "Precision < +-10 %". The temperature and length of storage (in days) are displayed in tables 11 through 19. As apparent in table 17 OXCBZ seems unstable at room temperature. And all other analytes seem stable for the length of time examined. Graphs of the course of stability of OXCBZ can be found in appendix I.

Table 11, Stability Levetiracetam

mg/L Location Mean CV (%) N Precision (%) days within limit

5.0 Room temperature 4.83 7.6 7 96.6 15 5.0 + 4 °C 4.88 5.9 7 97.6 15 5.0 - 20 °C 4.90 6.7 6 97.9 15 50.0 Room temperature 47.82 12.1 7 95.6 15 50.0 + 4 °C 48.87 9.5 7 97.7 15 50.0 - 20 °C 50.23 7.9 7 100.5 15

Table 12, Stability Phenobarbital

Table 13, Stability Lamotrigine

(27)

Table 14, Stability CBZ-Diol

Table 15, Stability 10-OH-CBZ

Table 16, Stability CBZ-Epoxide

(28)

Table 17, Stability OXCBZ

Table 18, Stability Phenytoin

3.0 Room temperature 3.01 6.0 7 100.2 15 3.0 + 4 °C 3.06 8.2 7 102.1 15 3.0 - 20 °C 2.99 6.7 6 99.7 15 30.0 Room temperature 29.94 19.7 7 99.8 15 30.0 + 4 °C 29.40 12.9 7 98.0 15 30.0 - 20 °C 27.56 6.6 7 91.9 15 Table 19, Stability CBZ

During evaporation phase of sample preparation, If samples are left in heating block for an extended period of time (>5 min), CV rises above 15% (n=4). If samples are not redissolved in mobile phase within 15 minutes, CV rises above 15% (n=4).

(29)

4.5 Inter- and intraday accuracy and precision

The degree of scatter inter- and intraday and precision thereof, was determined and is apparent in table 20. Prerequisites were accuracy<15% and precision <15%.

Table 20, Inter- and intraday accuracy and precision

Interday Intraday n Mean CV (%) Precision (%) n Mean CV (%) Precision (%) Levetiracetam Low 6 5.16 0.7 103.2 6 49.34 7.0 98.7 High 6 49.53 5.6 99.1 6 47.53 2.3 95.1 Phenobarbital Low 6 4.92 1.4 98.4 6 2.08 13.8 104.1 High 6 48.04 1.9 96.1 6 19.53 5.5 97.7 Lamotrigine Low 6 1.84 2.2 91.9 6 1.02 3.6 101.7 High 6 20.08 4.4 100.4 6 10.25 0.9 102.5 CBZ-Diol Low 6 1.01 0.6 101.0 6 4.93 6.4 98.7 High 6 10.13 3.0 101.3 6 48.29 1.9 96.6 10-OH-CBZ Low 6 4.93 1.3 98.6 6 4.93 6.4 98.7 High 6 47.83 1.8 95.7 6 48.29 1.9 96.6 CBZ-Epoxide Low 6 0.48 1.3 96.0 6 0.53 12.0 105.3 High 6 4.95 2.7 99.0 6 4.89 4.8 97.8 OXCBZ Low 6 0.95 1.4 94.7 6 0.95 8.1 94.6 High 6 9.67 2.8 96.7 6 9.10 7.8 91.0 Phenytoin Low 6 2.85 1.4 95.1 6 2.99 6.7 99.7 High 6 28.57 5.8 95.2 6 27.99 5.6 93.3 CBZ Low 6 1.90 1.2 95.2 6 1.98 6.3 99.0 High 6 19.51 2.9 97.6 6 19.34 4.0 96.7

(30)

4.6 Recovery

The recovery validation result is displayed in table 21.

Table 21, Recovery of analytes including test for outliers.

Levetiracetam HET (IS) Phenobarbital Lamotrigine CBZ-Diol 10-OH-CBZ

Low QC - extract Average (mg/L) 21.5 72.1 111.2 60.9 19.3 185.2 SD 1.2 2.1 3.3 1.6 0.6 6.0 CV (%) _5.7 _3.0 _2.9 _2.7 _3.0 _3.2 n= 8 8 8 8 8 8 Suspect 19.4 75.8 115.9 63.5 18.5 195.1 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 2.126 calculated G 1.707 1.729 1.432 1.560 1.410 1.644 Reject ? No No No No No No Low QC - 100% Average (mg/L) 22.2 72.1 78.7 45.8 27.2 133.5 SD 4.7 4.4 16.9 10.5 5.9 29.6 CV (%) _11.3% _6.3% _10.5% _10.9% _10.3% _10.5% n= 8 8 8 8 8 8 Suspect 12.3 78.5 42.2 22.8 14.3 68.7 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 2.126 calculated G 2.107 1.462 2.164 2.185 2.185 2.187

Reject? No No Yes Yes Yes Yes

Low QC Recovery (%) after rejecting outlier

91.1% 100.7% 132.5% 124.3% 66.6% 129.7% High QC - Extract Average (mg/L) 195.9 66.7 922.0 555.7 181.4 1506.4 SD 13.4 6.2 86.2 54.3 17.3 132.9 CV (%) _6.8% _9.3% _9.4% _9.8% _9.6% _8.8% n= 8 8 8 8 8 8 Suspect 258.9 98.1 1352.2 820.2 271.2 2204.8 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 2.126 calculated G 2.163 2.198 2.191 2.180 2.209 2.215

Reject? Yes Yes Yes Yes Yes Yes

High QC - Extract Average (mg/L) 215.1 70.8 734.0 461.5 260.1 1227.7 SD 10.6 4.0 39.5 26.6 14.0 64.9 CV (%) _4.9% _5.6% _5.4% _5.8% _5.4% _5.3% n= 8 8 8 8 8 8 Suspect 235.6 78.6 803.9 510.6 285.1 1344.8 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 2.126 calculated G 1.933 1.951 1.770 1.849 1.784 1.805 Reject? No No No No No No

High QC - Recovery (%) after rejecting outlier

(31)

Table 21, continuation recovery of analytes including test for outliers.

CBZ-Epoxide OXCBZ HEX (IS) Phenytoin CBZ

Low QC - extract Average (mg/L) 26.3 25.0 124.6 83.4 81.6 SD 0.9 0.8 4.8 3.1 3.1 CV (%) _3.2% _3.2% _3.8% _3.7% _3.8% n= 8 8 8 8 8 Suspect 27.6 26.1 132.9 88.2 86.7 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 calculated G 1.492 1.384 1.741 1.551 1.680 Reject ? No No No No No Low QC - 100% Average (mg/L) 19.0 18.0 91.0 57.9 56.4 SD 4.4 4.0 5.4 13.1 12.7 CV (%) _11.6% _10.4% _6.1% _10.3% _10.3% n= 8 8 8 8 8 Suspect 9.5 9.1 99.5 29.1 28.3 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 calculated G 2.150 2.199 1.577 2.205 2.208

Reject? Yes Yes No Yes Yes

91.1% 100.7% 132.5% 124.3% 66.6% High QC - Extract Average (mg/L) 248.1 246.8 121.8 679.0 745.0 SD 47.7 49.2 19.7 78.1 147.7 CV (%) _9.5% _9.9% _8.1% _10.7% _9.6% n= 8 8 8 8 8 Suspect 354.6 356.9 165.5 801.8 1076.7 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 calculated G 2.234 2.237 2.224 1.573 2.246

Reject? Yes Yes Yes No Yes

High QC - Extract Average (mg/L) 184.4 181.1 95.3 538.5 537.4 SD 10.3 10.8 6.2 29.7 31.0 CV (%) _5.6% _6.0% _6.5% _5.5% _5.8% n= 8 8 8 8 8 Suspect 202.2 200.4 106.7 593.7 594.2 Critical G (n=8) 2.126 2.126 2.126 2.126 2.126 calculated G 1.722 1.794 1.834 1.859 1.832 Reject? No No No No No

(32)

4.7 Patient samples

Previously measured patient samples are subjected to analysis and the concentrations compared are displayed in table 22.

Table 22, Patient samples measured with new method compared to counter method in

mg/L Analyte New method (mg/L) Counter method

(mg/L) Deviation Counter method

patient 1 CBZ 10.23 11.06 92% HPLC-DAD

patient 2 Lamotrigine 3.22 3.92 82% HPLC-DAD

patient 3 Lamotrigine 4.61 4.91 94% HPLC-DAD

patient 5 Phenobarbital 1 1 100% Immunoassay

patient 6 Levetiracetam 21.2 25.4 83% Outsourced: HPLC-DAD

Lamotrigine 4.3 4.4 98% HPLC-DAD

Phenytoin 18 23 78% Immunoassay

patient 8 Levetiracetam 31.7 41.3 77% Outsourced: HPLC-DAD

Lamotrigine 2 2.2 91% HPLC-DAD

patient 9 CBZ 9 11 82% HPLC-DAD

patient 10 Phenytoin 4.9 5.5 89% Immunoassay

patient 11 Levetiracetam 20 24.2 83% Outsourced: HPLC-DAD

Lamotrigine 3.7 3.5 106% HPLC-DAD

patient 12 Phenobarbital 22 23 96% Immunoassay

Lamotrigine 2.33 2.27 103% HPLC-DAD CBZ 6.1 7.3 84% HPLC-DAD 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -50 -20 0 20 40 60 80 100 120 140 160 180 200 220 240 260 280

30002072013 LOQ+patient #17 [modif ied by s dekoning] 05820 UV _V IS_1 mA U min 1 - L evetirac etam - 4,4 47 2 - 4 ,770 3 - b -hydrox y-ethy l theop hilline (IS) -

5,803 4 - L amotrig ine - 10 ,8405 - C BZ-Diol - 11, 853 6 - C BZ-ep oxide - 12,7 13 7 - H exoba rbital (IS) - 14 ,097 8 - C arbamaz epine - 15, 507 WV L:210 nm

Figure 6, A chromatogram of a patients serum sample, receiving CBZ, Lamotrigine and

(33)

When there is not a sufficient amount of sample volume available; it may be possible to use less sample and add drug free serum to add up to the total of 200 µl. Table 23 shows the precision achieved with this method of sample preperation.

Table 23,

Analyte patient serum

(µL) drug free serum (µL) Measured (mg/L) Expected (mg/L) Precision (%)

Levetiraceteam 50 150 5.35 4.89 109.4 100 100 10.64 9.78 108.8 200 0 19.56 - - Phenobarbital 50 150 5.69 5.46 104.2 100 100 10.96 10.92 100.4 200 0 21.84 - - Lamotrigine 50 150 0.82 0.58 140.8 100 100 1.32 1.17 113.3 200 0 2.33 - - CBZ-Diol 50 150 1.19 1.03 115.8 100 100 2.16 2.06 105.1 200 0 4.11 - - CBZ-Epoxide 50 150 0.34 0.32 107.1 100 100 0.63 0.64 99.2 200 0 1.27 - - Phenytoin 50 150 2.40 2.43 99.0 100 100 4.77 4.86 98.1 200 0 9.73 - - CBZ 50 150 1.60 1.51 105.7 100 100 3.04 3.02 100.7 200 0 6.05 - -

(34)

4.8 Selectivity

Possible interfering substances were analyzed by sampling external quality controls from the groups of benzodiazepines, cardiac medication and painkillers. Figure 7A is amiodaron, desethylamiodaron, coffeïn, digoxine, flecaïnide and theofylline, figure 7B is clobazam, desmethylclobazam, clonazepam, diazepam, desmethyldiazepam, oxazepam and temazepam and figure 7C is ibuprofen, paracetamol, salicylzuur.

A 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -100 -50 0 50 100 150 200 250 300 350 400 450 500 550 600 650

70003072013 loq pat v alid rec etc #47 KKGT CA RD UV _V IS_1

mA U min 1 - 5,040 WV L:210 nm B 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -20 -10 0 10 20 30 40 50 60 70 80 90 100 110

12003072013 loq pat v alid rec etc #46 [modif ied by s dekoning] KKGT Benz od UV _V IS_1 mA U min 1 - 15,9 13 2 - 16,6 80 3 - 17,2 47 WV L:210 nm C 0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -100 -50 0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750

80003072013 loq pat v alid rec etc #45 [modif ied by s dekoning] KKGT PIJN UV _V IS_1 mA U

min

1 - 16,3 40

WV L:210 nm

Figure 7ABC, Chromtograms of testing for interferences in chromatography 7A is cardiac

(35)

4.9 Carryover

Carry over was examined by injecting the highest calibration standard followed by a drug free serum sample spiked with only the internal standards (blank).

A -0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,3 -5,1 -2,0 0,0 2,0 4,0 6,0 8,0 10,0 12,0 14,0 16,0 18,0 20,0 22,0 24,0 26,0 28,0 30,0

32,701072013 HBH LOQ #7 [modif ied by s dekoning] blanc o 2012C UV _V IS_1 mA U

min

1 - 4,797 2 - b-hy droxy-e

thyl theop hilline (IS) -

5,843 3 - H exobarbita l (IS) - 14,127 WV L:210 nm Az B -0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -12,4 -10,0 -7,5 -5,0 -2,5 0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5 25,0 27,5 30,0

33,101072013 HBH LOQ #10 [modif ied by s dekoning] s tandaard 1/100 UV _V IS_1 mA U

min

1 - Lev etiracetam

- 4,504 2 - 4,807 3 - b-hy droxy-e thyl theop hilline (IS) - 5,854 4 - Feno barbital - 10,220 5 - 10,9 60 6 - 10-OH-C BZ - 12, 507 7 - CBZ-ep oxide - 12,7 97 8 - Oxca rbazepi ne - 13,597 9 - Hex obarbita l (IS) - 14 ,110 10 - Fen ytoïne - 111 -4,8 Carba54 maze pine - 15,570 WV L:210 nm Az

Figure 8AB, A: Carry over (drug free serum spiked only with IS), compared to B: 1/100th_of

the lowest calibration standard.

For reference: 1/100th_{of the lowest calibration standard results in the following}

concentrations: Levetiracetam 0.05 mg/L, Phenobarbital 0.05 mg/L, Lamotrigine 0.02 mg/L CBZ-Diol 0.01 mg/L, 10-OH-CBZ 0.05 mg/L, CBZ-Epoxide 0.005 mg/L, OXCBZ 0.01 mg/L, Phenytoin 0.03 mg/L and CBZ 0.02 mg/L.

(36)

5 Discussion

There are several methods reported for the detection of Levetiracetam. Greiner-Sosanko et

al. (2007) reported in the Journal of Chromatographic Science a method that was able to

detect Levetiracetam with use of gas chromatography and a nitrous phosphor detector. As there were no possibilities of measuring other analytes with use of this method and the higher preference to develope a method based on liquid chromatography with diode array detector this was not an option. Matar (2008) reported a method in the Journal of Pharmaceutical and Biomedical analysis for the analysis of Levetiracetam in human plasma with liquid chromatography-tandem mass spectrometry. As there was no room on the mass spectrometers at the department this was also not an option. However Oláh et al. (2012) reported a method in the journal of Chromatographic Science that suggested that Levetiracetam can be measure with photodiode absorbance. Wavelength was set at 210 nm and the system was equipped with a C18 column. Antonilli et al. (2011) reported a thin layer chromatography method. Juenke et al. (2011) reported an ultra pressure liquid chromatography method couple with tandem mass spectrometry detector. During the research it became apparent that the method of choice was liquid chromatography coupled with mass spectrometry, especially for multi drug assays.

The method however developed seems fit for its purpose, that is measuring serum levels of patients receiving Levetiracetam, Phenobarbital, Lamotrigine, OXCBZ, Phenytoin and CBZ. Although it is been demonstrated possible to measure Phenobarbital and Phenytoin levels in serum, the main method of analysis used within LUMC is an immunoassay. Though if, for whatever reason, the method of analysis by immunoassay may not be possible, the method described in this report can be utilized. Current applied methods of analysis for Lamotrigine, OXCBZ and CBZ can however be replaced by this method.

As Levetiracetam does not show a significant absorbance spectrum (appendix II) due to the lack of chromophores, the detection wavelength is set to a less specific 210 nm. Previously reported methods of analysis utilized the detector at wavelengths ranging from 205 to 220. [25-27]

(37)

Although it may seem that the 10-OH-CBZ and CBZ-Epoxide peaks are not separated adequately (figure 5), the tails of the analytes do not interfere with the centre (top) of the peaks. Given the fact that this method utilizes peak height to calculate concentrations and that CBZ-Epoxide and 10-OH-CBZ are metabolites of respectively CBZ and OXCBZ which are not co administered to patients, it may therefore be noted that it is indeed adequately separated. If need be (e.g. CBZ-Epoxide level is extremely low, and must be analyzed for inter-laboratory comparison), it is possible to separate CBZ-Epoxide and 10-OH-CBZ by changing eluens bottle A (chapter 3.5) from '100% methanol' to '150/900 acetonitrile/methanol'. The downside is that Levetiracetam suffers resolution from matrix peaks and can therefore not be quantified under these adapted conditions.

It is also possible to measure high concentrations in patient samples, e.g. overdose cases, of Levetiracetam, Phenobarbital, Lamotrigine, OXCBZ, Phenytoin and CBZ (table 8). CBZ-Diol, a metabolite of CBZ, is however limited to quantification at a maximum 10 mg/L. The lower limit of quantification (table 9) levels were determined and showed no abnormalities. The matrix effects were minimal (table 10) and all analytes passed the test. However GPO was not a suitable (cheaper) replacement due to interferences in chromatography at retention times overlapping CBZ-Diol, 10-OH-CBZ and CBX-Epoxide. (appendix III).

The inter- and intraday accuracy and precision show plausible results all within the required 15%. Note must be taken to check the needle in the auto analyzer for air bubbles, as this will decrease your precision and especially accuracy.

Although the stability test is not yet finished, the stability of storage of OXCBZ at fridge temperature (4 ˚C) and even worse at room temperature is not adequate enough. However during the same amount of time stored at -20 ˚C the progress of the storage showed little decline and mostly general scatter (appendix) which can be explained by the inter day accuracy (table 20). Earlier reported storage methods of preference have all pointed towards storage of at least -20 ˚C. Therefore OXCBZ patient samples should be frozen as soon as possible or in any event straight after receiving sample at the department. [28] It was also found important during sample preparation to not leave the sample in the heating block at 40 ˚C for an unnecessary extended period of time (15 mins, n=4).

(38)

The recovery, in contrast to the inter- and intraday precision, shows significant deviations to what is expected. Phenobarbital, Lamotrigine, 10-OH-CBZ, HEX (IS) and Phenytoin display a degree of recovery that is thought to be incorrect (120-130% recovery). This test has however been redone numerous times, and generates approximately the same values. Notice that the accuracy is generally higher in the low QC samples compared to the high QC samples. This may be the case due to the fact that for the high QC two actions are conducted (2x 10 µL WS) and for the lower QC one action is required with regards to the WS (1x 2µL). At first it was thought that there had been a mistake in spiking the 100% samples in mobile phase. Recovery was redone, and same problem still existed. The following idea was that it may be a dissolving problem; so the 100% sample was sonified for 5 minutes. Same problem still existed. Mobile phase was increased in amount of methanol to ease the dissolving of analytes, but same results were obtained. However with increasing the mobile phase (from 10 to 50%) methanol volume, bad peak formation was observed. So the area was determined instead of height, but the same result was achieved. During every analysis the blank is analyzed after the highest standard, to be able to determine whether a carryover effect has taken place. The carryover is however comparable to the response of 1/100th_{of the lowest standard and thus lies far under the lower limit of}

quantification.

It was also found that if there was not a sufficient volume of patient sample available for analysis, it could be diluted with drug-free serum. When diluted 50/50 (sample/drug free serum), an average precision of 103.7 ± 5.6% was achieved. When diluting it 25/75, an average precision of 106.9 ± 5.6% was achieved. To be certain of this fact, it should be conducted more than twice. Lamotrigine seemed unsuitable for diluting, therefore it has been taken out of the equation. The unsuitability can however be explained by the LLOQ for Lamotrigine. The 25/75 dilution was performed and concentration was afterwards expected to be under the determined LLOQ level.

(39)

6 Conclusion

This projects aim of priority was to develop a method for routine analysis of Levetiracetam in serum. In conclusion to this aim, it may be said that the method is suitable for routine analysis. Furthermore this method also allows for the quantitative analysis of eight other common anti epileptic drugs and metabolites of which six can from now on be determined with this method and for the other two this method will serve as a back-up for the currently applied immunoassay methods.

(40)

7 Recommendations

- The problem with recovery still needs to be tackled. As it is an important part of the validation. Although the recovery seems consistent, a recovery yield of 130% seems impossible.

- The storage stability validation is still in progress and will develop more results regarding the stability in the three different sample storage temperatures.

(41)

Literature

1. Jickells, S.N., A, Clarke’s Analytical Forensic Toxicology. 3 ed. 2008: Pharmaceutical Press. 672.

2. Stroke, N.I.o.N.D.a. NINDS Epilepsy Information Page. 2013 [cited 2013 05-03]; Available from: http://www.ninds.nih.gov/disorders/epilepsy/epilepsy.htm.

3. Vincent Iannelli, M.D. 2011 August 13, 2011 [cited 2013 04-03]; Available from:

http://pediatrics.about.com/cs/conditions/a/msp_epilepsy.htm.

4. Beghi, E., et al., Recommendation for a definition of acute symptomatic seizure. Epilepsia, 2010. 51(4): p. 671-5.

5. Patsalos, P.N. and J.W. Sander, Newer antiepileptic drugs. Towards an improved

risk-benefit ratio. Drug Saf, 1994. 11(1): p. 37-67.

6. Robert S. Fisher, M.D., Ph.D. and M.S.M.P.o. Neurology. What Causes Epilepsy. [cited 2013 06-03]; Available from:

http://neurology.stanford.edu/divisions/e_05.html.

7. Aur, D., Understanding the physical mechanism of transition to epileptic seizures. J Neurosci Methods, 2011. 200(1): p. 80-5.

8. Kids, F.H.M.G.-E.c.f. Symptomatic Epilepsy. [cited 2013 06-03]; Available from:

http://www.epilepsycenterforkids.com/services-and-specialties/conditions/symptomatic-epilepsy.

9. MD, C.N. What are epileptic lesions and what causes them? - NEUROLOGY EXPERT

FORUM. 2000 [cited 2013 06-03]; Available from:

http://www.medhelp.org/posts/Neurology/What-are-epileptic-lesions-and-what-causes-them/show/293471.

10. Juan G Ochoa, M.W.R., MD. Antiepileptic Drugs. 2012 25-10-2012 [cited 2013 07-03]; Available from:

http://emedicine.medscape.com/article/1187334-overview#aw2aab6b3.

11. Robert S. Fisher, M.D., Ph.D. SUMMARY OF ANTIEPILEPTIC DRUGS. 2009 [cited 2013 06-03]; Available from:

http://www.epilepsy.com/epilepsy/newsletter/sept09/aeds.

12. Surges, R., K.E. Volynski, and M.C. Walker, Is levetiracetam different from other

antiepileptic drugs? Levetiracetam and its cellular mechanism of action in epilepsy revisited. Ther Adv Neurol Disord, 2008. 1(1): p. 13-24.

13. Gower, A.J., et al., ucb L059, a novel anti-convulsant drug: pharmacological profile

in animals. Eur J Pharmacol, 1992. 222(2-3): p. 193-203.

14. Klitgaard, H., Levetiracetam: the preclinical profile of a new class of antiepileptic

drugs? Epilepsia, 2001. 42 Suppl 4: p. 13-8.

15. Jain, K.K., An assessment of levetiracetam as an anti-epileptic drug. Expert Opin Investig Drugs, 2000. 9(7): p. 1611-24.

16. agency, E.m. EMEA - Keppra. 2013 [cited 2013 07-03]; Available from:

http://www.ema.europa.eu/ema/index.jsp?curl=pages/medicines/human/medicines/

000277/human_med_000872.jsp&murl=menus/medicines/medicines.jsp.

17. Ulloa, C.M., A. Towfigh, and J. Safdieh, Review of levetiracetam, with a focus on the

extended release formulation, as adjuvant therapy in controlling partial-onset seizures. Neuropsychiatr Dis Treat, 2009. 5: p. 467-76.

18. De Smedt, T., et al., Levetiracetam: part II, the clinical profile of a novel

anticonvulsant drug. CNS Drug Rev, 2007. 13(1): p. 57-78.

19. Patsalos, P.N., Pharmacokinetic profile of levetiracetam: toward ideal characteristics. Pharmacol Ther, 2000. 85(2): p. 77-85.

(42)

20. Pellock, J.M., et al., Pharmacokinetic study of levetiracetam in children. Epilepsia, 2001. 42(12): p. 1574-9.

21. Patsalos, P.N., et al., In situ metabolism of levetiracetam in blood of patients with

epilepsy. Epilepsia, 2006. 47(11): p. 1818-21.

22. Harris, D.C., Quantitative Chemical Analysis. 5th ed. 2007, New York: Craig Bleyer. 23. Administration, U.S.D.o.H.a.H.S.F.a.D., C.f.D.E.a. Research, and C.f.V. Medicine.

Bioanalytical Method Validation. 2001 [cited 2013 17-03]; Available from:

http://www.fda.gov/downloads/Drugs/.../Guidances/ucm070107.pdf.

24. Centre, L.U.M., Valideren van HPLC- en GC-analysemethoden. 2004, LUMC: Clinical Pharmacy and Toxicology. p. 4.

25. Contin M, Mohamed S, Albani F, Riva R, Brauzzi A. Simple and validated HPLC-UV analysis of levetiracetam in deproteinized plasma of patients with epilepsy. J Chromatogr B Analyt Technol Biomed Life Sci. 2008;873:129–132.

26 Juenke J, Brown PI, Urry FM, McMillin GA. Drug monitoring and toxicology: a procedure for the monitoring of levetiracetam and zonisamide by HPLC-UV. J Anal Toxicol. 2006;30:27–30

27 Martens-Lobenhoffer J, Bode-Böger SM. Determination of levetiracetam in human plasma with minimal sample pretreatment. J Chromatogr B Analyt Technol Biomed Life Sci. 2005;819:197–200.

28 R. V. S. Nirog et al, Quantification of Oxcarbazepine and Its Active Metabolite 10-Hydroxycarbazepine in Human Plasma by High-performance Liquid Chromatography. Arzneimittelforschung Drug Research, 2006. 56(7): p. 517-523

(43)

Appendix I

Storage stability of OXCBZ

(44)

Apendix II

(45)

Apendix III

Blank GPO spiked with 10 µL.

0,0 1,0 2,0 3,0 4,0 5,0 6,0 7,0 8,0 9,0 10,0 11,0 12,0 13,0 14,0 15,0 16,0 17,0 18,0 -20 0 20 40 60 80 100 120 140 160 180 200 220

25022052013_1 matrix onderzoek valid #8 GPO 0 UV_VIS_1 mAU min 1 - b -hydrox y-ethy l theop hilline (IS) - 5,88 7 2 - 1 1,813 3 - 1 2,270 4 - H exoba rbita l (IS ) - 14 ,173 WVL:210 nm

Development, optimization and validation of a HPLC DAD method assaying Levetiracetam and common AED in serum

Final report

Sdffdfsdf

Development, optimization and validation of a HPLC DAD

method assaying Levetiracetam and common AED in serum.

Final report

Sdffdfsdf

Development, optimization and validation of a HPLC DAD

method assaying Levetiracetam and common AED in serum.

Acknowledgements

Table of Contents

1 Abstract

2 Introduction

2.1 Pharmacy

2.1.1 Academic hospital pharmacy

2.1.2 Clinical toxicology

2.2 Epilepsy

2.2.1 Idiopathic epilepsy

2.2.2 Symptomatic epilepsy

2.3 Antiepileptic drugs

2.3.1 Calcium channel blockers

2.3.2 Sodium channel blockers

2.4 Levetiracetam

2.4.1 Mechanism of action

2.4.2 Pharmacokinetics and metabolism

2.5 Other antiepileptic drugs and metabolites of interest

2.6 Analytical techniques

2.6.1 Liquid/liquid extraction

2.6.2 High performance liquid chromatography

2.7 Method validation

2.8 Overall objective

3 Materials and Methods

3.1 Chemicals and Reagents

3.2 Calibration Standards and Quality Control Samples

3.3 Patient serum samples

3.4 Sample preparation

3.5 Instrumentation and Chromatographic Conditions.

3.6 Method validation

4 Results

4.1 Chromatography

4.2 Quantification range

4.3 Matrix effect

4.4 Stability

4.5 Inter- and intraday accuracy and precision

4.6 Recovery

4.7 Patient samples

4.8 Selectivity

4.9 Carryover

5 Discussion

6 Conclusion

7 Recommendations

Literature

Appendix I

Apendix II

Apendix III