Assessment of the correlations of lacosamide concentrations in saliva and serum in patients with epilepsy

Transcription

1 Accepted: 21 January 2018 DOI: /epi BRIEF COMMUNICATION Assessment of the correlations of lacosamide concentrations in saliva and serum in patients with epilepsy Christian Brandt 1 Christian G. Bien 1 Renate Helmer 2 Theodor W. May 2 1 Bethel Epilepsy Center, Mara Hospital, Bielefeld, Germany 2 Society for Epilepsy Research, Bethel Epilepsy Center, Bielefeld, Germany Correspondence Christian Brandt, Bethel Epilepsy Center, Mara Hospital, Bielefeld, Germany. christian.brandt@mara.de Funding information UCB Pharma (Germany, Monheim) Summary Therapeutic drug monitoring of antiepileptic drugs is based on patient serum samples. In this study, we evaluated the correlation between lacosamide (LCM) steady state concentrations in serum and saliva samples. Additionally, we investigated the relation with daily dose, and assessed the feasibility of saliva collection. This was an open-label, single center study including data from 25 patients at the Bethel Epilepsy Center treated with LCM ( mg/d). Samples were collected in the morning (fasting values) and in selected cases at 50 minutes to 5 hours after the morning dose. Nonsignificant differences in the mean LCM morning (trough) concentration in serum and saliva were observed. Serum and saliva concentrations across all samples were highly correlated, (r =.874), with a slightly lower correlation when only fasting values were analyzed (r =.860). Higher correlation with daily dosages was observed in serum samples (r =.773) than in saliva samples (r =.604). Serum and saliva concentrations increased significantly after intake of the LCM morning dose (P <.001). The median absolute and percentage increase of LCM in serum were moderately lower than in saliva samples, with a few outliers in saliva samples. Consequently, saliva could offer great clinical potential to monitor drug concentrations and guide LCM treatment in epileptic patients. KEYWORDS antiepileptic drugs, correlations, lacosamide, saliva, serum, steady state 1 INTRODUCTION Lacosamide (LCM; [R]-2-acetamido-N-benzyl-3-methoxypropionamide) is a third-generation antiepileptic drug (AED) approved in 2008 as adjuvant anticonvulsive treatment for adults with partial onset seizures, and more recently as monotherapy. 1 It exerts its therapeutic effect by assisting the gradual inactivation of voltage-gated sodium channels, thereby stabilizing hyperexcitable neuronal membranes and inhibiting emissions. 1 See the literature for pharmacokinetic details. 2 5 Therapeutic drug monitoring of AEDs assists in adjusting AED doses, while also avoiding potential associated toxicity. 5 8 In line with recommendations from the U.S. Food and Drug Administration, which support the use of noninvasive sampling methods, as well as the relatively easy collection, transport and storage of samples, saliva is currently being extensively considered for a wide range of applications (ie, to detect biomarkers, measure hormone/enzyme levels, or monitor the development of diseases). 9 The correlation found in the concentration of different drugs (LCM among others) in plasma/serum samples and saliva has led to the consideration of the potential of saliva as a source for pharmacokinetic studies. 3,4,10 12 The clinical use of saliva would be especially relevant for pediatric use, elderly or disabled patients, and those with circulatory/blood clotting disorders, where blood sampling may be challenging. 12 Previous investigations on healthy individuals and patients on LCM therapy have also shown a high and linear correlation between LCM concentrations in saliva and serum. 10,12 A recent study suggested that LCM serum e34 Wiley Periodicals, Inc International League Against Epilepsy wileyonlinelibrary.com/journal/epi Epilepsia. 2018;59:e34 e39.

2 BRANDT ET AL. e35 concentrations can be considered a valuable indicator of LCM concentrations in cerebrospinal fluid. 3 Due to the close correlation of LCM concentrations in serum and saliva, 10,12 saliva may also be an indicator of LCM in the central nervous system. However, the reported ratios of LCM concentrations in saliva compared to serum differed markedly. Whereas in a study by Greenaway et al. 10 the mean total LCM concentrations in serum were about 10- fold higher than LCM concentrations in saliva (ratio of concentrations of free LCM in serum to LCM in saliva: 1.27), Cawello et al. 12 reported that the mean LCM concentration in saliva was only about 10% higher than in serum. The aim of this study was to further investigate the correlation between saliva and serum drug concentrations in epileptic patients at steady state receiving LCM across a wide range of doses. 2 MATERIALS AND METHODS 2.1 Study design This study was an open-label, single center trial conducted using biological samples of 25 adult inpatients prescribed Vimpat (UCB Pharma, Monheim, Germany). All patients had been appropriately informed about the research project and provided written consent for the samples to be used in this research. The study plan was approved by the regional ethics committee. 2.2 Sample collection, storage, and analysis Saliva samples from LCM-treated subjects were taken in the morning to determine fasting values. Further samples were taken during the day when side effects were suspected. Subjects were asked to provide saliva samples (0.5-1 ml) by expectorating without stimulation into collection tubes. Samples were transferred into polypropylene tubes, which were sealed and stored at 20 C until use. Additionally, blood was drawn from subjects at the same time as saliva samples. In case of additional saliva/serum samples, the second sample was usually taken about 1 hour after the morning administration of LCM, based on the LCM median concentration peak. 12 Saliva and blood sampling, and LCM dosage, were recorded as follows: fasting values (before LCM morning intake) versus values taken after the morning dosing (range = 50 minutes to 5 hours). See Table S1 for the analytical method. 2.3 Statistical analysis Descriptive statistics including frequency and percentage values, mean, standard deviation (SD), variance, median, and 25% and 75% percentile were calculated. Linear regression analyses, Pearson correlation (r), Spearman rank coefficient (r s ), and nonparametric tests (Wilcoxon test) were used for statistical analyses. Twosided P values are reported. Daily LCM dose was adjusted for body weight or for approximated volume of distribution (V d ). The apparent V d for LCM equals that of total body water; therefore, common empiric equations for approximation of total body water were used. 13 For male subjects: V d ½LŠ ¼ 0:3625 weight [kg] þ 0:2239 height [cm] 0:1387 age [years] 14:47 For female subjects: V d [LŠ ¼ 0:2363 weight [kg] þ 0:1962 height [cm] 0:0272 age [years] 10:26 Because of the fluctuation of LCM concentration during the day described in previous studies, 4 linear regression analysis was conducted on 2 groups: (1) fasting samples of blood and saliva and (2) all samples (including nonfasting samples of blood and saliva). Statistical analysis was conducted using SPSS 21.0 (IBM, Armonk, NY, USA). 3 RESULTS Twenty-five subjects were included in the study. One subject was excluded because LCM concentrations in saliva were extremely unusual, indicating an error (possibly due to mixing up samples or inadequate sealing of tubes) during sample collection (see Table S1, Patient 22). Twenty-four fasting and 13 nonfasting paired saliva/serum specimens were included. The 24 subjects (aged years) received on average 381 mg LCM/d (range = mg/d); 2 subjects (8.3%) received LCM in 1 daily dose, 14 (58.3%) in 2, 7 (29.8%) in 3, and 1 patient (4.2%) in 4 daily doses. All samples were collected in LCM steady state (>2 days since last change of LCM dose). 3.1 Concentration of LCM in serum and saliva samples The mean concentration of LCM in all serum and saliva samples was lg/ml and lg/ml, respectively (mean SD), whereas fasting values were lg/ml (serum) and lg/ml (saliva). The differences between mean LCM concentrations in serum and saliva were not statistically significant (P =.056 for all values; P =.310 for fasting values only; 2-sided exact Wilcoxon test). This translated into a slightly higher

3 e36 BRANDT ET AL. average concentration in saliva (LCM Saliva :LCM Serum ): for all samples, for fasting values. There was a comparatively large standard deviation, caused by the presence of outliers (Figure 1). Individual data are presented for each subject in Table S1. A strong linear correlation between the concentration of LCM in saliva and serum was found when only fasting values or all samples were included, with r =.860 (95% confidence interval = , n = 24; r 2 =.739) for fasting values (Figure 1A) and r =.874 (95% confidence interval = , n = 37; r 2 =.764) for all samples (Figure 1B). Furthermore, there was no significant correlation between the daily LCM dose (adjusted for body weight, or for approximated volume of distribution, or not adjusted) and the ratio LCM Saliva /LCM Serum (all correlation A B LCM Saliva Concentration (μg/ml) LCM Saliva Concentration (μg/ml) 25 R 2 = LCM Serum Concentration (μg/ml) 25 R 2 = LCM Serum Concentration (μg/ml) FIGURE 1 Relationship between lacosamide (LCM) saliva and serum concentrations for A, fasting samples of 24 patients and B, all 37 samples of the 24 patients including nonfasting samples (n = 37) coefficients < 0.2, all P-values >.2); that is, the ratio LCM Saliva /LCM Serum was not dependent on LCM dose ( mg). 3.2 Increase of saliva and serum concentrations after intake of LCM The concentration of LCM before and after drug intake was compared in 12 paired saliva/serum samples. Table 1 and Figure 2A,B illustrate the absolute and percentage increase of LCM concentrations in serum and saliva after intake of the LCM morning dose (both P <.001, 2-sided exact Wilcoxon test). The (absolute and percentage) increase in LCM concentrations after intake did not differ significantly between serum and saliva (P =.092 and P =.052 for serum and saliva, respectively; 2-sided exact Wilcoxon test). Again, outliers were observed in the saliva values (Figure 2A,B). 3.3 Relationship between daily LCM doses and LCM concentration in serum and saliva Due to the expected fluctuations throughout the day, only fasting values were included in analysis of the correlation between LCM concentration in saliva or serum and daily dose per body weight (mg/kg). Correlations of LCM concentrations in serum and saliva with LCM dose per body weight were moderate and somewhat higher for LCM concentrations in serum (r =.690 and r =.578 for serum and saliva, respectively; P <.01, n = 24). However, the correlation coefficients were higher (r =.749 and r =.671 for serum and saliva, respectively; P <.01, n = 22) after exclusion of 2 patients coadministered enzyme-inducing AEDs (150 mg phenobarbital; 450 mg carbamazepine) and especially after adjustment of the LCM dose by the approximated volume of distribution (instead of adjustment by body weight; r =.875 and r =.757 for serum and saliva, respectively, P <.01, n = 22). 4 DISCUSSION Therapeutic drug monitoring (TDM) is an important tool in guiding and adjusting AED therapy. 6 Earlier studies on LCM have shown a high correlation of LCM concentrations in serum and saliva samples in patients and healthy individuals, although results varied between studies. 10,12 In this study, we further elucidate the correlation between the LCM concentration in serum and saliva across a wide range of dosages. Our data confirmed a high correlation between LCM concentrations in serum and saliva samples in patients. These observations support previous findings from Cawello

4 BRANDT ET AL. e37 TABLE 1 LCM concentrations in serum and saliva in morning (trough) samples and samples after intake of LCM morning dose n Mean Median SD Min Max Only morning (trough) samples, n = 24 samples of 24 patients LCM serum concentration, lg/ml LCM saliva concentration, lg/ml Ratio saliva:serum All samples (morning samples and samples after intake of LCM), n = 37 samples of 24 patients LCM serum concentration, lg/ml LCM saliva concentration, lg/ml Ratio saliva:serum Increase of LCM concentrations in serum and saliva after intake of LCM morning dose, n = 12 matching serum/saliva pairs of 12 patients LCM serum concentration, lg/ml Before intake of LCM After intake of LCM 9.75 a Increase, % LCM saliva concentration, lg/ml Before intake of LCM After intake of LCM 11.2 a Increase, % The conversion factor from lg/ml to l mol L 1 is 3.99 (ie, 1 lg/ml [mg/l] = 3.99 lmol L 1 ).LCM, lacosamide; SD, standard deviation. a Significantly higher LCM concentrations after intake of the LCM morning dose (P <.001, 2-sided exact Wilcoxon test; in Patient 10, only the first sample after the morning dose was considered, thus n = 12 samples were analyzed). et al., 12 although the correlation observed in their study (r 2 =.958 for all samples, or r 2 =.950 for >0.5 hours after LCM dosing samples) was higher than the one here described. This may be due to the Cawello data being determined from a clinical trial in healthy volunteers, whereas ours are derived from clinical routine. In addition, the potential existence of confounding factors in patients compared to healthy individuals might be a source of variability and error, possibly resulting in the outliers detected in saliva samples. 13 The correlation of concentrations of LCM in serum and saliva here described relates more closely to that found by Greenaway et al. 10 (r 2 =.842); however, these authors reported LCM serum concentrations to be 10-fold higher than those in plasma and protein binding to be 90%. In contrast, the present report indicates <15% protein-binding with salivary concentration similar to non protein-bound LCM. This finding agrees with all previous reports (see also Patsalos et al. 14 ). Because of the close linear relationship between LCM concentrations in serum and in saliva, the ratio saliva/serum may be used for transformations of saliva concentrations in serum concentrations (simply by dividing the saliva concentration by this ratio). Then, therapeutic ranges for LCM may be applied to these transformed LCM saliva concentrations. Our data showed no major differences in the concentration of LCM between saliva and serum. The increase of LCM concentrations after administration of the morning dose was slightly but not significantly greater for saliva samples than serum around 1 hour after dosing, similar to findings by Cawello et al. 12 The correlations between LCM concentrations and doses increased after exclusion of 2 patients with enzyme-inducing AEDs and especially after adjustment of the LCM dose by the approximated volume of distribution. This indicates that enzyme-inducing AEDs may affect the LCM concentration-dose relationship and that adjusting the LCM dose by approximated VD may better predict the concentration in serum and saliva than adjusting by body weight. We are aware of the limitations of our study. Difficulties with saliva collection or with handling (storage) of saliva samples may have caused the outliers, although this cannot be verified. Our investigations suggest that saliva is a suitable alternative biological sample for TDM. ACKNOWLEDGMENTS This study was funded by UCB Pharma (Monheim, Germany). We would like to thank Uwe J urgens (Pharmacological Laboratory, Society for Epilepsy Research, Bielefeld, Germany) for his valuable support in determining LCM concentrations in saliva and serum, and Willi Cawello (UCB) for valuable discussions. Medical writing support was provided by Jennifer Engelmoer (TransPerfect, Utrecht, The Netherlands), Alejandra Martinez de Pinillos Bayona (TransPerfect, London, UK), and Renee Entzminger (TransPerfect, New York, NY, USA).

5 e38 BRANDT ET AL. A B % Increase a er LCM morning dose Increase a er LCM morning dose (μg/ml) Serum Serum AUTHOR CONTRIBUTIONS All authors were involved in the concept and design of the study, analysis and interpretation of the data, critical review of the manuscript, and the decision to submit for publication. Data acquisition was performed by C.B., R.H., and T.W.M.; T.W.M. performed the statistical analysis. CONFLICT OF INTEREST Saliva Saliva FIGURE 2 A, Absolute and B, percentage increase in lacosamide (LCM) saliva and serum concentrations in most patients approximately 1 hour after intake of morning LCM dose in 12 patients. The numbered points mark outliers; the number corresponds to the patient ID (see Table S1) C.B. has received personal compensation from Actelion, Desitin, Eisai, Otsuka, Pfizer, SKS, UCB, and USL Pharma for serving on a scientific advisory board or for speaking activities or congress travel; his institution has received 16 financial support for research activities from Otsuka and UCB Pharma. C.G.B. gave scientific advice to Eisai and UCB; undertook industry-funded travel with the support of Eisai, UCB, Desitin, and Grifols; and obtained honoraria for speaking engagements from Eisai, UCB, Desitin, Diamed, Fresenius Medical Care, Biogen, and Euroimmun. He received research support from Diamed and Fresenius Medical Care. He is a consultant to the Laboratory Krone, Bad Salzuflen, Germany, regarding neural antibodies and therapeutic drug monitoring for antiepileptic drugs. R.H. received financial support from Eisai for visiting scientific meetings. T.W.M. received financial support from UCB and Desitin for visiting scientific meetings, and received honoraria for speaking engagements from Eisai and Desitin. We confirm that we have read the Journal s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. ORCID Christian Brandt REFERENCES 1. Errington AC, Coyne L, St ohr T, Selve N, Lees G. Seeking a mechanism of action for the novel anticonvulsant lacosamide. Neuropharmacology. 2006;50: Cawello W. Clinical pharmacokinetic and pharmacodynamic profile of lacosamide. Clin Pharmacokinet. 2015;54: May TW, Brandt C, Helmer R, et al. Comparison of lacosamide concentrations in cerebrospinal fluid and serum in patients with epilepsy. Epilepsia. 2015;56: Sattler A, Schaefer M, May TW, et al. Fluctuation of lacosamide serum concentrations during the day and occurrence of adverse drug reactions first clinical experience. Epilepsy Res. 2011;95: Svendsen T, Brodtkorb E, Baftiu A, et al. Therapeutic drug monitoring of lacosamide in Norway: focus on pharmacokinetic variability, efficacy and tolerability. Neurochem Res. 2017;42: Patsalos PN, Berry DJ, Bourgeois BF, et al. Antiepileptic drugs best practice guidelines for therapeutic drug monitoring: a position paper by the subcommission on therapeutic drug monitoring. ILAE Commission on Therapeutic Strategies. Epilepsia. 2008;49: Brandt C, Baumann P, Eckermann G, et al. Therapeutic drug monitoring in epileptology and psychiatry [in German]. Nervenarzt. 2008;79: Brandt C, May TW. Therapeutic drug monitoring of newer antiepileptic drugs. Laboratoriums Medizin. 2011;35: Aro K, Wei F, Wong DT, et al. Saliva liquid biopsy for point-ofcare applications. Front Public Health. 2017;5: Greenaway C, Ratnaraj N, Sander JW, et al. Saliva and serum lacosamide concentrations in patients with epilepsy. Epilepsia. 2011;52:

6 BRANDT ET AL. e Idkaidek N, Arafat T, Hamadi H, et al. Saliva versus plasma bioequivalence of azithromycin in humans: validation of class I drugs of the salivary excretion classification system. Drugs R D. 2017;17: Cawello W, B okens H, Nickel B, et al. Tolerability, pharmacokinetics, and bioequivalence of the tablet and syrup formulations of lacosamide in plasma, saliva, and urine: saliva as a surrogate of pharmacokinetics in the central compartment. Epilepsia. 2013;54: Fountain N, Staelens L, Tytgat D, et al. Low lacosamide plasma protein binding in lacosamide-na ıve patients. Neurology. 2012;78: P Patsalos PN, Berry DJ. Therapeutic drug monitoring of antiepileptic drugs by use of saliva. Ther Drug Monit. 2013;35:4 29. SUPPORTING INFORMATION Additional Supporting Information may be found online in the supporting information tab for this article. How to cite this article: Brandt C, Bien CG, Helmer R, May TW. Assessment of the correlations of lacosamide concentrations in saliva and serum in patients with epilepsy. Epilepsia. 2018;59:e34 e39.