CLB-09034; No. of pages: 4; 4C: Clinical Biochemistry xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
3Q3
Ilona Burianová a,b, Klára Bořecká a,⁎ a
6
a r t i c l e
7 8 9 10 11
Article history: Received 9 February 2015 Received in revised form 1 May 2015 Accepted 15 May 2015 Available online xxxx
12 13 14 15 16 17 18
Keywords: Antiepileptic drugs Carbamazepine Carbamazepine-10,11-epoxide High-performance liquid chromatography Immunoassay Therapeutic drug monitoring
i n f o
R O
Department of Clinical Biochemistry, Thomayer Hospital, Prague, Czech Republic Department of Biochemistry, Faculty of Science, Charles University, Prague, Czech Republic
a b s t r a c t
T
E
D
P
Objectives: Carbamazepine (CBZ) is primarily used in the treatment of epilepsy as well as trigeminal neuralgia. It is metabolized by the liver to its pharmacologically active metabolite carbamazepine-10,11-epoxide (CBZ-E), which is potentially toxic. The aim of our study was to measure CBZ and CBZ-E in our patient set and to consider the introduction of CBZ-E to routine therapeutic drug monitoring (TDM). High-performance liquid chromatography (HPLC) and Chemiluminescence Microparticle Immuno Assay (CMIA) methods for the measurement of CBZ were compared. Design and Methods: We simultaneously measured serum concentrations of CBZ and CBZ-E using HPLC. Serum concentrations of CBZ were also analysed by CMIA. For the measurements we chose patients (ages 5–67 years) on monotherapy (n = 51), patients taking CBZ with antiepileptic drugs (AEDs) susceptible to pharmacokinetic interaction including phenytoin, phenobarbital, primidone and valproic acid (n = 56) and patients taking other AEDs (n = 44). Results: Patient's serum levels of CBZ-E ranged from 1.38–27.79 μmol/L with a mean value of 6.96 ± 4.07 μmol/L . The CBZ-E/CBZ ratio increased significantly in patients taking phenytoin, phenobarbital, primidone and valproic acid. CBZ concentrations measured by CMIA were lower than those obtained by HPLC (mean difference of 3.8 μmol/L). The Passing Bablok regression showed acceptable agreement between these two methods. Conclusions: Based on our results, we do not consider the introduction of the active metabolite of CBZ to routine TDM to be necessary. However, it might be beneficial in patients taking CBZ with AEDs susceptible to pharmacokinetic interaction to avoid any potential adverse effects. A close correlation between CMIA and HPLC method was found for the measurement of CBZ serum concentrations. © 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
R
E
C
b
O
4 5
F
2
Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?
1Q2
R
38 39 43 41 40 42
Introduction
45
Carbamazepine (CBZ) is used in the treatment of epilepsy, bipolar and psychotic disorder and was introduced in the treatment of trigeminal neuralgia [1]. CBZ is metabolized in the liver by the cytochrome P450 (CYP) isoforms, mainly CYP3A4, to its pharmacologically active metabolite carbamazepine-10,11-epoxide (CBZ-E) which is further metabolized by epoxide hydrolase to an inactive trans-carbamazepine diol [2]. Because CBZ affects the pharmacokinetics of other antiepileptic drugs (AEDs) and vice versa, leading to both increase and decrease of CBZ levels, and due to the fact that there is a weak correlation between dose and serum levels, especially in the elderly, the therapeutic drug monitoring (TDM) of CBZ is strongly recommended [3,4]. Well defined therapeutic and toxic ranges also support the need of TDM of CBZ.
48 49 50 51 52 53 54 55 56
U
46 47
N C O
44
⁎ Corresponding author at: Department of Clinical Biochemistry, Thomayer Hospital, Vídeňská 800, 140 59 Prague 4, Czech Republic. Fax: +420 261082435. E-mail address: [email protected] (K. Bořecká).
The CBZ-E contributes to clinical response, but it is not routinely monitored because there is no available immunoassay and its measurement is limited to chromatographic techniques [5,6]. According to some studies, co-administration of other anticonvulsants could increase CBZE concentrations and the CBZ-E/CBZ ratio [7–9]. Serum CBZ concentrations are usually within the therapeutic ranges, but patients could have some symptoms of toxicity, probably caused by high levels of the metabolite. The CBZ-E/CBZ ratios significantly increase in patients with moderate to severe renal disease [7]. Increase in this ratio was also observed in infants and young children mainly due to higher rates of metabolism [10]. Therefore, monitoring of CBZ-E could be useful for epileptic patients to decrease the risk of toxicity. Therapeutic and toxic levels of CBZ are 4–12 mg/L (17–51 μmol/L) and N 20 mg/L (N 84.6 μmol/L), respectively [11,12]. The therapeutic range for CBZ-E is not well established, but is considered to be 0.4–4 mg/L (1.6–16 μmol/L). To prevent the toxic effect of epoxide, its concentration should not exceed 9 mg/L (35.6 μmol/L) [13]. The aim of our study was to measure CBZ-E in patients undergoing routine control of CBZ levels in our laboratory and consider a possible
http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014 0009-9120/© 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
19 20 21 22 23 24 25 26 27 28 Q4 29 30 Q5 31 32 33 34 35 36 37
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
86 87
100 Q6
For the study the patients undergoing routine control of the AEDs levels were divided into three groups. The first group comprises patients on monotherapy (n = 51). The second group consisted of patients taking CBZ and other AEDs including clonazepam, lacosamide, lamotrigine, levetiracetam, topiramate and zonisamide (Polytherapy I; n = 44). In the third group were patients taking CBZ with the drugs which are highly susceptible to pharmacokinetic interactions including phenytoin, phenobarbital, primidone and valproic acid (Polytherapy II; n = 56). According to the completed request form the indications for TDM in our patient groups were periodic monitoring, therapy optimization, ineffective drug treatment and suspected non-compliance. No suspected intoxication was the reason for the measurement of CBZ levels. All patients (ages 5–67 years) had been taking CBZ for more than three months meaning that blood samples were collected at steady-state plasma concentration.
101
CBZ-E and CBZ analysis
102
111 112
Blood was collected into blood collection tubes without any gel separator and anticoagulants. After centrifugation, serum concentrations of CBZ were determined by CMIA using Architect® i 2000SR analyser (Abbott, Prague, Czech Republic). The rest of the sample was stored at − 24 °C until HPLC analysis. Serum levels of CBZ and CBZ-E were measured simultaneously on Shimadzu High Performance Liquid Chromatograph LC-10VP series (Shimadzu, Kyoto, Japan) using HPLC reagent kit for Antiepileptic Drugs in serum/plasma (Chromsystems Instruments and Chemicals GmbH, Munich, Germany). Sample preparation and analysis were performed according to the manufacturer's instructions.
113
Data evaluation
114 115 116 117
Data were evaluated using MedCalc statistical software version 13 (MedCalc®, Mariakerke, Belgium). The results were analysed by the box and whisker plots, Passing Bablok regression, Bland–Altman plots, Spearman's correlation and the Mann–Whitney test.
118
Results
119
Monitoring of the CBZ-E and CBZ
120
In our patient group (n = 151), concentrations of CBZ were within the therapeutic ranges except for nine (six above and three below). Average concentrations of CBZ and CBZ-E were 33.35 ± 12.95 μmol/L and 6.96 ± 4.07 μmol/L, respectively. In only one patient, CBZ-E concentration nearly reached the toxic level (the parent drug was within the range). The patient was taking valproic acid and did not suffer from any side effects. From six cases of CBZ overdosing, CBZ-E exceeded therapeutic levels in two of them. No significant differences in CBZ-E concentrations between single patient groups were found (Fig. 1). As shown in Fig. 2, the results did not show any significant increase in CBZE/CBZ ratio in patients taking concomitant AEDs (Polytherapy I), compared to patients on monotherapy. When phenytoin, phenobarbital,
107 108 109 110
121 122 123 124 125 126 127 128 129 130 131
C
105 106
E
103 104
R
98 99
primidone and valproic acid were included (Polytherapy II), the ratio of CBZ-E to parent drug significantly increased in comparison to both groups. We also compared CBZ-E/CBZ ratios in 13 patients taking CBZ and lamotrigine with patients on monotherapy, but did not find any significant differences (data not shown).
132 133 134 135 136
HPLC and CMIA method comparison
137
As we expected, the serum CBZ concentrations measured by HPLC were higher than by CMIA. The mean value of the difference (bias) was 3.8 μmol/L (11.8%) and 95% limit of agreement ranged from −2.5 to 10.1 μmol/L as shown in Bland–Altman plot (Fig. 3). The Passing Bablok regression analysis revealed close agreement between these two methods (Fig. 4). Spearman's rank correlation coefficient 0.949 (p b 0.0001) also revealed a high correlation between CBZ measured by CMIA and HPLC.
138 139 140 141 142 143 144 Q7 145
R
96 97
O
94 95
Fig. 1. Box plot of CBZ-E concentrations of patients on monotherapy, polytherapy including clonazepam, lacosamide, lamotrigine, levetiracetam, topiramate and zonisamide (Polytherapy I) and patients taking CBZ with phenytoin, phenobarbital, primidone and valproic acid (Polytherapy II). No significant differences were found. The solid horizontal line represents the upper limit of therapeutic ranges.
C
92 93
N
90 91
U
88 89
F
Participants
O
85
82
R O
Patients and methods
80 81
P
84
78 79
T
83
introduction of CBZ-E to routine therapeutic monitoring. We wanted to find out if there are any CBZ-E concentrations exceeding therapeutic or even toxic levels. We also focused on comparison of the relative proportion of CBZ-E and CBZ in epileptic patients on monotherapy and patients with concomitant antiepileptic treatment. Because introduction of CBZE to routine TDM would require replacement of commercial immunoassay with HPLC, we compared the CMIA with HPLC for the determination of CBZ in serum.
D
76 77
I. Burianová, K. Bořecká / Clinical Biochemistry xxx (2015) xxx–xxx
E
2
Fig. 2. Differences in CBZ-E/CBZ ratios in patients on monotherapy and polytherapy including clonazepam, lacosamide, lamotrigine, levetiracetam, topiramate and zonisamide (Polytherapy I). When phenytoin, phenobarbital, primidone and valproic acid were included (Polytherapy II), CBZ-E/CBZ ratios significantly increased which was verified by Mann–Whitney test (p b 0.001). Q1
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
I. Burianová, K. Bořecká / Clinical Biochemistry xxx (2015) xxx–xxx
3
R O
O
F
concentration reaching 57 μmol/L [14]. The other published results related to correlation between epoxide concentrations and toxicity are conflicting meaning that patients usually respond differently to treatment which is mainly attributed to genetic factors. The previous studies showed the increase in CBZ-E/CBZ ratio when CBZ is combined with other anticonvulsants including phenytoin, phenobarbital, primidone and valproic acid which is in agreement with our results [8,9,15]. Phenytoin, phenobarbital and primidone are wellknown cytochrome P450 inducers compared to valproic acid which inhibits CYP isoforms as well as epoxide hydrolase, the enzyme responsible for conversion of CBZ-E to an inactive form. They all can affect the metabolism of CBZ [16,17]. Most of the above cited publications mention the potential risk of toxicity when CBZ-E/CBZ ratio increases, but did not prove this; moreover, according to Theodore et al, there is only weak correlation between CBZ-E, CBZ levels and toxicity scores in patients on monotherapy and patients taking other AEDs as well [18]. Although this was reported, we proved the pharmacokinetic interactions leading to increase in CBZ-E/CBZ ratio and therefore keeping CBZ concentrations on the lower limit of therapeutic ranges could decrease any potential risk of side effects in patients taking CBZ with other enzyme inducers. When we compared the patients on monotherapy with patients taking other AEDs (Polytherapy I) we did not find any significant differences in CBZ-E/CBZ ratios. Many of these AEDs belong to the group of new AEDs which are considered to be less susceptible to pharmacokinetics interactions [17]. The effect of lamotrigine on CBZ-E/CBZ ratio was reported in several studies with conflicting results [19,20]. Lamotrigine is not an enzyme inducer or inhibitor but could be affected by other drugs including carbamazepine, eslicarbazepine, valproic acid etc. [16,17]. Our data did not reveal any significant differences between patients taking lamotrigine and patients on monotherapy, thus these observations support the papers suggesting no effect of lamotrigine on CBZ or CBZ-E. In general, TDM of the antiepileptic drugs is necessary in patients with renal insufficiency or hepatic disease, however, renal insufficiency as well as dialysis has no significant effect on plasma levels of carbamazepine. Concerning hepatic function, only severe hepatic disease could affect plasma levels and thus require TDM [21,22]. For that reason, we did not focus on patients with renal and liver diseases; moreover, biochemical analysis did not indicate any. According to analysis using Passing Bablok regression and Spearman's rank correlation, the similarity between CBZ concentrations measured by HPLC and CMIA was determined. HPLC methods show high sensitivity, specificity and accuracy and are considered as a gold standard for many assays. They are not influenced by cross-reaction or other interferences which is the common problem of some immunoassays [23]. The other problem of immunoassays is the matrix itself, which is usually eliminated in HPLC analysis. However, immunoassays are widely used in biochemical laboratories because they are sensitive, fast and easy to perform and do not require any sample preparation or further training to operate. Although it is evident that concentrations measured by HPLC were higher than by CMIA, we do not consider the bias with a mean difference of 3.8 μmol/L to be clinically important. The same results were also found in other studies where HPLC and immunoassay were compared [24]. In conclusion, according to our results and available data there is no need to monitor the active metabolite of CBZ in each patient undergoing routine control of CBZ levels. However, monitoring of the active metabolite of CBZ could be beneficial in patients taking CBZ with AEDs susceptible to pharmacokinetics interactions in case of developing adverse reactions when CBZ is in therapeutic ranges. In case of introduction of CBZ-E to routine therapeutic monitoring, concentrations slightly exceeding the upper limit of therapeutic ranges should not always be the reason for reduction in a dose of CBZ due to the fact the relation of CBZ-E and toxicity has not been clearly demonstrated. The results should be interpreted critically according to clinical state of patient and response to therapy.
According to our results, it is still open to debate, whether CBZ-E should be introduced to routine therapeutic monitoring. When we exclude the patients with CBZ overdosing, CBZ-E exceeded the therapeutic ranges in only two cases and its average concentration was very low. In only one patient, the epoxide concentration nearly reached the toxic levels. The patient was taking CBZ and valproic acid and did not suffer from any liver or kidney disease. It is also necessary to emphasize the fact that therapeutic and toxic concentrations of CBZ-E were not well established, because CBZ-E is not routinely measured. Because CBZ-E contributes to clinical response, in some clinical laboratories we can find therapeutic ranges defined as the sum of CBZ + CBZ-E b 51 μmol/L which correspond to the upper limit of therapeutic ranges for CBZ. It follows that concentrations of CBZ reaching the upper limit of therapeutic ranges could automatically represent an increased risk of side effects due to the contribution of CBZ-E. However, Tomson et al. reported that there are no subjective side effects in patients treated by CBZ-E even in plasma
159 160 161 162
C
E
157 158
R
155 156
R
153 154
N C O
151 152
U
149 150
D
147 148
E
Discussion
T
146
P
Fig. 3. Bland Altman plots of the difference between CBZ concentrations measured by HPLC and CMIA. The solid line represents the bias and 95% confidence limits for the bias are displayed as dashed horizontal lines.
Fig. 4. The Passing Bablok regression analyses of HPLC and CMIA methods for the measurement of CBZ. Regression line equation was: y = 1.303 + 0.839x; 95% confidence intervals (CI) for intercept −0.009 to 2.510 and for slope 0.800 to 0.878 indicated good agreement. Cusum test for linearity indicated no significant deviation from linearity. Residual standard deviation (RSD) was 1.958 with ± 1.96 RSD interval ranged from −3.838 to 3.838.
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
Authors are without any conflict of interest.
234
Contributions
235
238
IB and KB contributed to the conception and design of the study. IB performed the analysis and interpreted the data. IB and KB contributed to drafting the article or revising it critically for important intellectual content. KB gave final approval of the version to be submitted.
239
Acknowledgements
240 241
246
We would like to thank Alena Linhartová (Department of Clinical Pharmacy, Thomayer Hospital, Prague, Czech Republic) for the specialist advice and Vladimír Królikowski (BioTech a.s., Prague, Czech Republic) for the additional support. The author (IB) thanks to the Ministry of Education of the Czech Republic for the scholarship and to Specific University Research Project No. 33779266 awarded by Charles University Prague for additional financial support.
247
References
R
R
E
C
T
[1] Blom S. Trigeminal neuralgia: its treatment with a new anticonvulsant drug (G-32883). Lancet 1962;1:839–40. [2] Kerr BM, Thummel KE, Wurden CJ, Klein SM, Kroetz DL, Gonzalez FJ, et al. Human liver carbamazepine metabolism. Role of CYP3A4 and CYP2C8 in 10,11-epoxide formation. Biochem Pharmacol 1994;47:1969–79. [3] Bondareva IB, Jelliffe RW, Gusev EI, et al. Population pharmacokinetic modelling of carbamazepine in epileptic elderly patients: implications for dosage. J Clin Pharm Ther 2006;31:211–21. [4] Neels HM, Sierens AC, Naelaerts K, Scharpé SL, Hatfield GM, Lambert WE. Therapeutic drug monitoring of old and newer anti-epileptic drugs. Clin Chem Lab Med 2004; 42:1228–55. [5] Oh Eun kyung, Ban Eunmi, Woo Jong Soo, Kim Chong-Kook. Analysis of carbamazepine and its active metabolite, carbamazepine-10,11-epoxide, in human plasma using high-performance liquid chromatography. Anal Bioanal Chem 2006;386: 1931–6.
O
248 249 250 251 252 253 Q8 254 255 256 257 258 259 Q9 260 261 262
C
244 245
N
242 243
U
236 237
F
233
O
Conflict of interest statement
R O
232
[6] Leite CE, Petersen GO, Lunardelli A, Thiesen FV. A high-performance liquid chromatography method for the determination of carbamazepine and carbamazepine10,11-epoxide and its comparison with chemiluminescent immunoassay. Clin Chem Lab Med 2009;47:458–63. [7] Tutor-Crespo MJ, Hermida J, Tutor JC. Relative proportions of serum carbamazepine and its pharmacologically active 10, 11-epoxy derivative: effect of polytherapy and renal insufficiency. Ups J Med Sci 2008;113:171–80. [8] Brodie MJ, Forrest G, Rappeport WG. Carbamazepine 10, 11 epoxide concentrations in epileptics on carbamazepine alone and in combination with other anticonvulsants. Br J Clin Pharmacol 1983;16:747–50. [9] Potter JM, Donnelly A. Carbamazepine-10,11-epoxide in therapeutic drug monitoring. Ther Drug Monit 1998;20:652–7. [10] Pynnönen S, Sillanpää M, Frey H, Iisalo E. Carbamazepine and its 10,11-epoxide in children and adults with epilepsy. Eur J Clin Pharmacol 1997;11:129–33. [11] Schulz M, Iwersen-Bergmann S, Andresen H, Schmoldt A. Therapeutic and toxic blood concentrations of nearly 1000 drugs and other xenobiotics. Crit Care 2012; 16:136. [12] Hiemke C, Baumann P, Bergemann N, Conca A, Dietmaier O, Egberts K, et al. AGNP consensus guidelines for therapeutic drug monitoring in psychiatry: update 2011. Pharmacopsychiatry 2011;44:195–235. [13] Shorvon Simon D. Handbook of epilepsy treatment3rd ed. Wiley-Blackwell978-14051-9818-9; December 2010[436 pp., (1405198184)]. [14] Tomson T, Almkvist O, Nilsson BY, Svensson JO, Bertilsson L. Carbamazepine-10,11epoxide in epilepsy a pilot study. Arch Neurol 1990;47:888–92. [15] See comment in PubMed Commons belowAltafullah I, Talwar D, Loewenson R, Olson K, Lockman LA. Factors influencing serum levels of carbamazepine and carbamazepine-10,11-epoxide in children. Epilepsy Res 1989;4:72–80. [16] Spina E, Pisani F, Perucca E. Clinically significant pharmacokinetic drug interactions with carbamazepine. An update. Clin Pharmacokinet 1996;31:198–214. [17] Johannessen SI, Johannessen Landmark C. Antiepileptic drug interactions — principles and clinical implications. Curr Neuropharmacol 2010;8:254–67. [18] Theodore WH, Narang PK, Holmes MD, Reeves P, Nice FJ. Carbamazepine and its epoxide: relation of plasma levels to toxicity and seizure control. Ann Neurol 1989;25:194–6. [19] Schapel GT, Dollman W, Beran RG, Dunagan FM. No effect of LTG on CBZ and CBZ-E concentrations. Epilepsia 1991;32(Suppl. 1):S58. [20] Gidal BE, Rutecki P, Shaw R, Maly MM, Collins DM, Pitterle ME. Effect of lamotrigine on carbamazepine epoxide:carbamazepine serum concentration ratios in adult patients with epilepsy. Epilepsy Res 1997;28:207–11. [21] Kandrotas RJ, Oles KS, Gal P, Love JM. Carbamazepine clearance in hemodialysis and hemoperfusion. DICP 1989;23:137–40. [22] Asconapé JJ. Use of antiepileptic drugs in hepatic and renal disease. Handb Clin Neurol 2014;119:417–32. [23] Ismail AAA. Interference from endogenous antibodies in automated immunoassays: what laboratorians need to know. J Clin Pathol 2009;62:673–9. [24] Krasniqi S, Zeitlinger M, Bauer S. Comparison between the immunoassay and high performance liquid chromatography for therapeutic monitoring of carbamazepine and phenytoine. Biochem Med 2010;20:341–9.
P
231
The results of method comparison showed a close correlation between CMIA and HPLC and we do not consider the calculated bias to be clinically important.
D
229 230
I. Burianová, K. Bořecká / Clinical Biochemistry xxx (2015) xxx–xxx
E
4
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 Q10 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
Contents lists available at ScienceDirect
Clinical Biochemistry journal homepage: www.elsevier.com/locate/clinbiochem
3Q3
Ilona Burianová a,b, Klára Bořecká a,⁎ a
6
a r t i c l e
7 8 9 10 11
Article history: Received 9 February 2015 Received in revised form 1 May 2015 Accepted 15 May 2015 Available online xxxx
12 13 14 15 16 17 18
Keywords: Antiepileptic drugs Carbamazepine Carbamazepine-10,11-epoxide High-performance liquid chromatography Immunoassay Therapeutic drug monitoring
i n f o
R O
Department of Clinical Biochemistry, Thomayer Hospital, Prague, Czech Republic Department of Biochemistry, Faculty of Science, Charles University, Prague, Czech Republic
a b s t r a c t
T
E
D
P
Objectives: Carbamazepine (CBZ) is primarily used in the treatment of epilepsy as well as trigeminal neuralgia. It is metabolized by the liver to its pharmacologically active metabolite carbamazepine-10,11-epoxide (CBZ-E), which is potentially toxic. The aim of our study was to measure CBZ and CBZ-E in our patient set and to consider the introduction of CBZ-E to routine therapeutic drug monitoring (TDM). High-performance liquid chromatography (HPLC) and Chemiluminescence Microparticle Immuno Assay (CMIA) methods for the measurement of CBZ were compared. Design and Methods: We simultaneously measured serum concentrations of CBZ and CBZ-E using HPLC. Serum concentrations of CBZ were also analysed by CMIA. For the measurements we chose patients (ages 5–67 years) on monotherapy (n = 51), patients taking CBZ with antiepileptic drugs (AEDs) susceptible to pharmacokinetic interaction including phenytoin, phenobarbital, primidone and valproic acid (n = 56) and patients taking other AEDs (n = 44). Results: Patient's serum levels of CBZ-E ranged from 1.38–27.79 μmol/L with a mean value of 6.96 ± 4.07 μmol/L . The CBZ-E/CBZ ratio increased significantly in patients taking phenytoin, phenobarbital, primidone and valproic acid. CBZ concentrations measured by CMIA were lower than those obtained by HPLC (mean difference of 3.8 μmol/L). The Passing Bablok regression showed acceptable agreement between these two methods. Conclusions: Based on our results, we do not consider the introduction of the active metabolite of CBZ to routine TDM to be necessary. However, it might be beneficial in patients taking CBZ with AEDs susceptible to pharmacokinetic interaction to avoid any potential adverse effects. A close correlation between CMIA and HPLC method was found for the measurement of CBZ serum concentrations. © 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
R
E
C
b
O
4 5
F
2
Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?
1Q2
R
38 39 43 41 40 42
Introduction
45
Carbamazepine (CBZ) is used in the treatment of epilepsy, bipolar and psychotic disorder and was introduced in the treatment of trigeminal neuralgia [1]. CBZ is metabolized in the liver by the cytochrome P450 (CYP) isoforms, mainly CYP3A4, to its pharmacologically active metabolite carbamazepine-10,11-epoxide (CBZ-E) which is further metabolized by epoxide hydrolase to an inactive trans-carbamazepine diol [2]. Because CBZ affects the pharmacokinetics of other antiepileptic drugs (AEDs) and vice versa, leading to both increase and decrease of CBZ levels, and due to the fact that there is a weak correlation between dose and serum levels, especially in the elderly, the therapeutic drug monitoring (TDM) of CBZ is strongly recommended [3,4]. Well defined therapeutic and toxic ranges also support the need of TDM of CBZ.
48 49 50 51 52 53 54 55 56
U
46 47
N C O
44
⁎ Corresponding author at: Department of Clinical Biochemistry, Thomayer Hospital, Vídeňská 800, 140 59 Prague 4, Czech Republic. Fax: +420 261082435. E-mail address: [email protected] (K. Bořecká).
The CBZ-E contributes to clinical response, but it is not routinely monitored because there is no available immunoassay and its measurement is limited to chromatographic techniques [5,6]. According to some studies, co-administration of other anticonvulsants could increase CBZE concentrations and the CBZ-E/CBZ ratio [7–9]. Serum CBZ concentrations are usually within the therapeutic ranges, but patients could have some symptoms of toxicity, probably caused by high levels of the metabolite. The CBZ-E/CBZ ratios significantly increase in patients with moderate to severe renal disease [7]. Increase in this ratio was also observed in infants and young children mainly due to higher rates of metabolism [10]. Therefore, monitoring of CBZ-E could be useful for epileptic patients to decrease the risk of toxicity. Therapeutic and toxic levels of CBZ are 4–12 mg/L (17–51 μmol/L) and N 20 mg/L (N 84.6 μmol/L), respectively [11,12]. The therapeutic range for CBZ-E is not well established, but is considered to be 0.4–4 mg/L (1.6–16 μmol/L). To prevent the toxic effect of epoxide, its concentration should not exceed 9 mg/L (35.6 μmol/L) [13]. The aim of our study was to measure CBZ-E in patients undergoing routine control of CBZ levels in our laboratory and consider a possible
http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014 0009-9120/© 2015 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved.
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
19 20 21 22 23 24 25 26 27 28 Q4 29 30 Q5 31 32 33 34 35 36 37
57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
86 87
100 Q6
For the study the patients undergoing routine control of the AEDs levels were divided into three groups. The first group comprises patients on monotherapy (n = 51). The second group consisted of patients taking CBZ and other AEDs including clonazepam, lacosamide, lamotrigine, levetiracetam, topiramate and zonisamide (Polytherapy I; n = 44). In the third group were patients taking CBZ with the drugs which are highly susceptible to pharmacokinetic interactions including phenytoin, phenobarbital, primidone and valproic acid (Polytherapy II; n = 56). According to the completed request form the indications for TDM in our patient groups were periodic monitoring, therapy optimization, ineffective drug treatment and suspected non-compliance. No suspected intoxication was the reason for the measurement of CBZ levels. All patients (ages 5–67 years) had been taking CBZ for more than three months meaning that blood samples were collected at steady-state plasma concentration.
101
CBZ-E and CBZ analysis
102
111 112
Blood was collected into blood collection tubes without any gel separator and anticoagulants. After centrifugation, serum concentrations of CBZ were determined by CMIA using Architect® i 2000SR analyser (Abbott, Prague, Czech Republic). The rest of the sample was stored at − 24 °C until HPLC analysis. Serum levels of CBZ and CBZ-E were measured simultaneously on Shimadzu High Performance Liquid Chromatograph LC-10VP series (Shimadzu, Kyoto, Japan) using HPLC reagent kit for Antiepileptic Drugs in serum/plasma (Chromsystems Instruments and Chemicals GmbH, Munich, Germany). Sample preparation and analysis were performed according to the manufacturer's instructions.
113
Data evaluation
114 115 116 117
Data were evaluated using MedCalc statistical software version 13 (MedCalc®, Mariakerke, Belgium). The results were analysed by the box and whisker plots, Passing Bablok regression, Bland–Altman plots, Spearman's correlation and the Mann–Whitney test.
118
Results
119
Monitoring of the CBZ-E and CBZ
120
In our patient group (n = 151), concentrations of CBZ were within the therapeutic ranges except for nine (six above and three below). Average concentrations of CBZ and CBZ-E were 33.35 ± 12.95 μmol/L and 6.96 ± 4.07 μmol/L, respectively. In only one patient, CBZ-E concentration nearly reached the toxic level (the parent drug was within the range). The patient was taking valproic acid and did not suffer from any side effects. From six cases of CBZ overdosing, CBZ-E exceeded therapeutic levels in two of them. No significant differences in CBZ-E concentrations between single patient groups were found (Fig. 1). As shown in Fig. 2, the results did not show any significant increase in CBZE/CBZ ratio in patients taking concomitant AEDs (Polytherapy I), compared to patients on monotherapy. When phenytoin, phenobarbital,
107 108 109 110
121 122 123 124 125 126 127 128 129 130 131
C
105 106
E
103 104
R
98 99
primidone and valproic acid were included (Polytherapy II), the ratio of CBZ-E to parent drug significantly increased in comparison to both groups. We also compared CBZ-E/CBZ ratios in 13 patients taking CBZ and lamotrigine with patients on monotherapy, but did not find any significant differences (data not shown).
132 133 134 135 136
HPLC and CMIA method comparison
137
As we expected, the serum CBZ concentrations measured by HPLC were higher than by CMIA. The mean value of the difference (bias) was 3.8 μmol/L (11.8%) and 95% limit of agreement ranged from −2.5 to 10.1 μmol/L as shown in Bland–Altman plot (Fig. 3). The Passing Bablok regression analysis revealed close agreement between these two methods (Fig. 4). Spearman's rank correlation coefficient 0.949 (p b 0.0001) also revealed a high correlation between CBZ measured by CMIA and HPLC.
138 139 140 141 142 143 144 Q7 145
R
96 97
O
94 95
Fig. 1. Box plot of CBZ-E concentrations of patients on monotherapy, polytherapy including clonazepam, lacosamide, lamotrigine, levetiracetam, topiramate and zonisamide (Polytherapy I) and patients taking CBZ with phenytoin, phenobarbital, primidone and valproic acid (Polytherapy II). No significant differences were found. The solid horizontal line represents the upper limit of therapeutic ranges.
C
92 93
N
90 91
U
88 89
F
Participants
O
85
82
R O
Patients and methods
80 81
P
84
78 79
T
83
introduction of CBZ-E to routine therapeutic monitoring. We wanted to find out if there are any CBZ-E concentrations exceeding therapeutic or even toxic levels. We also focused on comparison of the relative proportion of CBZ-E and CBZ in epileptic patients on monotherapy and patients with concomitant antiepileptic treatment. Because introduction of CBZE to routine TDM would require replacement of commercial immunoassay with HPLC, we compared the CMIA with HPLC for the determination of CBZ in serum.
D
76 77
I. Burianová, K. Bořecká / Clinical Biochemistry xxx (2015) xxx–xxx
E
2
Fig. 2. Differences in CBZ-E/CBZ ratios in patients on monotherapy and polytherapy including clonazepam, lacosamide, lamotrigine, levetiracetam, topiramate and zonisamide (Polytherapy I). When phenytoin, phenobarbital, primidone and valproic acid were included (Polytherapy II), CBZ-E/CBZ ratios significantly increased which was verified by Mann–Whitney test (p b 0.001). Q1
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
I. Burianová, K. Bořecká / Clinical Biochemistry xxx (2015) xxx–xxx
3
R O
O
F
concentration reaching 57 μmol/L [14]. The other published results related to correlation between epoxide concentrations and toxicity are conflicting meaning that patients usually respond differently to treatment which is mainly attributed to genetic factors. The previous studies showed the increase in CBZ-E/CBZ ratio when CBZ is combined with other anticonvulsants including phenytoin, phenobarbital, primidone and valproic acid which is in agreement with our results [8,9,15]. Phenytoin, phenobarbital and primidone are wellknown cytochrome P450 inducers compared to valproic acid which inhibits CYP isoforms as well as epoxide hydrolase, the enzyme responsible for conversion of CBZ-E to an inactive form. They all can affect the metabolism of CBZ [16,17]. Most of the above cited publications mention the potential risk of toxicity when CBZ-E/CBZ ratio increases, but did not prove this; moreover, according to Theodore et al, there is only weak correlation between CBZ-E, CBZ levels and toxicity scores in patients on monotherapy and patients taking other AEDs as well [18]. Although this was reported, we proved the pharmacokinetic interactions leading to increase in CBZ-E/CBZ ratio and therefore keeping CBZ concentrations on the lower limit of therapeutic ranges could decrease any potential risk of side effects in patients taking CBZ with other enzyme inducers. When we compared the patients on monotherapy with patients taking other AEDs (Polytherapy I) we did not find any significant differences in CBZ-E/CBZ ratios. Many of these AEDs belong to the group of new AEDs which are considered to be less susceptible to pharmacokinetics interactions [17]. The effect of lamotrigine on CBZ-E/CBZ ratio was reported in several studies with conflicting results [19,20]. Lamotrigine is not an enzyme inducer or inhibitor but could be affected by other drugs including carbamazepine, eslicarbazepine, valproic acid etc. [16,17]. Our data did not reveal any significant differences between patients taking lamotrigine and patients on monotherapy, thus these observations support the papers suggesting no effect of lamotrigine on CBZ or CBZ-E. In general, TDM of the antiepileptic drugs is necessary in patients with renal insufficiency or hepatic disease, however, renal insufficiency as well as dialysis has no significant effect on plasma levels of carbamazepine. Concerning hepatic function, only severe hepatic disease could affect plasma levels and thus require TDM [21,22]. For that reason, we did not focus on patients with renal and liver diseases; moreover, biochemical analysis did not indicate any. According to analysis using Passing Bablok regression and Spearman's rank correlation, the similarity between CBZ concentrations measured by HPLC and CMIA was determined. HPLC methods show high sensitivity, specificity and accuracy and are considered as a gold standard for many assays. They are not influenced by cross-reaction or other interferences which is the common problem of some immunoassays [23]. The other problem of immunoassays is the matrix itself, which is usually eliminated in HPLC analysis. However, immunoassays are widely used in biochemical laboratories because they are sensitive, fast and easy to perform and do not require any sample preparation or further training to operate. Although it is evident that concentrations measured by HPLC were higher than by CMIA, we do not consider the bias with a mean difference of 3.8 μmol/L to be clinically important. The same results were also found in other studies where HPLC and immunoassay were compared [24]. In conclusion, according to our results and available data there is no need to monitor the active metabolite of CBZ in each patient undergoing routine control of CBZ levels. However, monitoring of the active metabolite of CBZ could be beneficial in patients taking CBZ with AEDs susceptible to pharmacokinetics interactions in case of developing adverse reactions when CBZ is in therapeutic ranges. In case of introduction of CBZ-E to routine therapeutic monitoring, concentrations slightly exceeding the upper limit of therapeutic ranges should not always be the reason for reduction in a dose of CBZ due to the fact the relation of CBZ-E and toxicity has not been clearly demonstrated. The results should be interpreted critically according to clinical state of patient and response to therapy.
According to our results, it is still open to debate, whether CBZ-E should be introduced to routine therapeutic monitoring. When we exclude the patients with CBZ overdosing, CBZ-E exceeded the therapeutic ranges in only two cases and its average concentration was very low. In only one patient, the epoxide concentration nearly reached the toxic levels. The patient was taking CBZ and valproic acid and did not suffer from any liver or kidney disease. It is also necessary to emphasize the fact that therapeutic and toxic concentrations of CBZ-E were not well established, because CBZ-E is not routinely measured. Because CBZ-E contributes to clinical response, in some clinical laboratories we can find therapeutic ranges defined as the sum of CBZ + CBZ-E b 51 μmol/L which correspond to the upper limit of therapeutic ranges for CBZ. It follows that concentrations of CBZ reaching the upper limit of therapeutic ranges could automatically represent an increased risk of side effects due to the contribution of CBZ-E. However, Tomson et al. reported that there are no subjective side effects in patients treated by CBZ-E even in plasma
159 160 161 162
C
E
157 158
R
155 156
R
153 154
N C O
151 152
U
149 150
D
147 148
E
Discussion
T
146
P
Fig. 3. Bland Altman plots of the difference between CBZ concentrations measured by HPLC and CMIA. The solid line represents the bias and 95% confidence limits for the bias are displayed as dashed horizontal lines.
Fig. 4. The Passing Bablok regression analyses of HPLC and CMIA methods for the measurement of CBZ. Regression line equation was: y = 1.303 + 0.839x; 95% confidence intervals (CI) for intercept −0.009 to 2.510 and for slope 0.800 to 0.878 indicated good agreement. Cusum test for linearity indicated no significant deviation from linearity. Residual standard deviation (RSD) was 1.958 with ± 1.96 RSD interval ranged from −3.838 to 3.838.
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228
Authors are without any conflict of interest.
234
Contributions
235
238
IB and KB contributed to the conception and design of the study. IB performed the analysis and interpreted the data. IB and KB contributed to drafting the article or revising it critically for important intellectual content. KB gave final approval of the version to be submitted.
239
Acknowledgements
240 241
246
We would like to thank Alena Linhartová (Department of Clinical Pharmacy, Thomayer Hospital, Prague, Czech Republic) for the specialist advice and Vladimír Królikowski (BioTech a.s., Prague, Czech Republic) for the additional support. The author (IB) thanks to the Ministry of Education of the Czech Republic for the scholarship and to Specific University Research Project No. 33779266 awarded by Charles University Prague for additional financial support.
247
References
R
R
E
C
T
[1] Blom S. Trigeminal neuralgia: its treatment with a new anticonvulsant drug (G-32883). Lancet 1962;1:839–40. [2] Kerr BM, Thummel KE, Wurden CJ, Klein SM, Kroetz DL, Gonzalez FJ, et al. Human liver carbamazepine metabolism. Role of CYP3A4 and CYP2C8 in 10,11-epoxide formation. Biochem Pharmacol 1994;47:1969–79. [3] Bondareva IB, Jelliffe RW, Gusev EI, et al. Population pharmacokinetic modelling of carbamazepine in epileptic elderly patients: implications for dosage. J Clin Pharm Ther 2006;31:211–21. [4] Neels HM, Sierens AC, Naelaerts K, Scharpé SL, Hatfield GM, Lambert WE. Therapeutic drug monitoring of old and newer anti-epileptic drugs. Clin Chem Lab Med 2004; 42:1228–55. [5] Oh Eun kyung, Ban Eunmi, Woo Jong Soo, Kim Chong-Kook. Analysis of carbamazepine and its active metabolite, carbamazepine-10,11-epoxide, in human plasma using high-performance liquid chromatography. Anal Bioanal Chem 2006;386: 1931–6.
O
248 249 250 251 252 253 Q8 254 255 256 257 258 259 Q9 260 261 262
C
244 245
N
242 243
U
236 237
F
233
O
Conflict of interest statement
R O
232
[6] Leite CE, Petersen GO, Lunardelli A, Thiesen FV. A high-performance liquid chromatography method for the determination of carbamazepine and carbamazepine10,11-epoxide and its comparison with chemiluminescent immunoassay. Clin Chem Lab Med 2009;47:458–63. [7] Tutor-Crespo MJ, Hermida J, Tutor JC. Relative proportions of serum carbamazepine and its pharmacologically active 10, 11-epoxy derivative: effect of polytherapy and renal insufficiency. Ups J Med Sci 2008;113:171–80. [8] Brodie MJ, Forrest G, Rappeport WG. Carbamazepine 10, 11 epoxide concentrations in epileptics on carbamazepine alone and in combination with other anticonvulsants. Br J Clin Pharmacol 1983;16:747–50. [9] Potter JM, Donnelly A. Carbamazepine-10,11-epoxide in therapeutic drug monitoring. Ther Drug Monit 1998;20:652–7. [10] Pynnönen S, Sillanpää M, Frey H, Iisalo E. Carbamazepine and its 10,11-epoxide in children and adults with epilepsy. Eur J Clin Pharmacol 1997;11:129–33. [11] Schulz M, Iwersen-Bergmann S, Andresen H, Schmoldt A. Therapeutic and toxic blood concentrations of nearly 1000 drugs and other xenobiotics. Crit Care 2012; 16:136. [12] Hiemke C, Baumann P, Bergemann N, Conca A, Dietmaier O, Egberts K, et al. AGNP consensus guidelines for therapeutic drug monitoring in psychiatry: update 2011. Pharmacopsychiatry 2011;44:195–235. [13] Shorvon Simon D. Handbook of epilepsy treatment3rd ed. Wiley-Blackwell978-14051-9818-9; December 2010[436 pp., (1405198184)]. [14] Tomson T, Almkvist O, Nilsson BY, Svensson JO, Bertilsson L. Carbamazepine-10,11epoxide in epilepsy a pilot study. Arch Neurol 1990;47:888–92. [15] See comment in PubMed Commons belowAltafullah I, Talwar D, Loewenson R, Olson K, Lockman LA. Factors influencing serum levels of carbamazepine and carbamazepine-10,11-epoxide in children. Epilepsy Res 1989;4:72–80. [16] Spina E, Pisani F, Perucca E. Clinically significant pharmacokinetic drug interactions with carbamazepine. An update. Clin Pharmacokinet 1996;31:198–214. [17] Johannessen SI, Johannessen Landmark C. Antiepileptic drug interactions — principles and clinical implications. Curr Neuropharmacol 2010;8:254–67. [18] Theodore WH, Narang PK, Holmes MD, Reeves P, Nice FJ. Carbamazepine and its epoxide: relation of plasma levels to toxicity and seizure control. Ann Neurol 1989;25:194–6. [19] Schapel GT, Dollman W, Beran RG, Dunagan FM. No effect of LTG on CBZ and CBZ-E concentrations. Epilepsia 1991;32(Suppl. 1):S58. [20] Gidal BE, Rutecki P, Shaw R, Maly MM, Collins DM, Pitterle ME. Effect of lamotrigine on carbamazepine epoxide:carbamazepine serum concentration ratios in adult patients with epilepsy. Epilepsy Res 1997;28:207–11. [21] Kandrotas RJ, Oles KS, Gal P, Love JM. Carbamazepine clearance in hemodialysis and hemoperfusion. DICP 1989;23:137–40. [22] Asconapé JJ. Use of antiepileptic drugs in hepatic and renal disease. Handb Clin Neurol 2014;119:417–32. [23] Ismail AAA. Interference from endogenous antibodies in automated immunoassays: what laboratorians need to know. J Clin Pathol 2009;62:673–9. [24] Krasniqi S, Zeitlinger M, Bauer S. Comparison between the immunoassay and high performance liquid chromatography for therapeutic monitoring of carbamazepine and phenytoine. Biochem Med 2010;20:341–9.
P
231
The results of method comparison showed a close correlation between CMIA and HPLC and we do not consider the calculated bias to be clinically important.
D
229 230
I. Burianová, K. Bořecká / Clinical Biochemistry xxx (2015) xxx–xxx
E
4
Please cite this article as: Burianová I, Bořecká K, Routine therapeutic monitoring of the active metabolite of carbamazepine: Is it really necessary?, Clin Biochem (2015), http://dx.doi.org/10.1016/j.clinbiochem.2015.05.014
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 Q10 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310
