Impact of Age and Concurrent Antiseizure Medication Use on Lacosamide Dose to Concentration Ratio and Dosing in Pediatric Patients

Introduction

Lacosamide (Vimpat; UCB, Smyrna, GA) is a highly effective antiseizure medication (ASM) that is labelled in the United States as monotherapy for patients ≥ 1 month with partial-onset seizures (i.e., focal seizures), and as adjunctive therapy for those ≥ 4 years with primary generalized tonic-clonic (i.e., genetic generalized epilepsies).¹ US Food and Drug Administration (FDA) approval was based primarily on extrapolation of efficacy data from adults with partial-onset seizures,² and was supported by limited efficacy and safety data in patients 1 month to < 17 years of age.^3–7

Lacosamide (LCM) is biotransformation by CYP2C9, CYP2C19, and CYP3A4.⁸ While all 3 isozymes appear within the first week of life they are not fully expressed until 6 months of age^9–11; hence, age could be an important variable in LCM dosing, especially in neonates and young infants.

LCM is used frequently as an adjunctive ASM for refractory epilepsy; hence, understanding the ability of other ASMs to influence its biodisposition is paramount to the effective management of epilepsy and avoidance of toxicity. Prescribing information notes that “the pharmacokinetics of LCM are similar when it is used as monotherapy or as adjunctive therapy.”¹ Although studies have investigated the impact of enzyme inducing and/or inhibiting ASM on LCM clearance, most reports involved healthy volunteers,^12,13 adults,^14–16 and older children.^14,17–19

The few studies investigating the influence of age and/or concomitant ASMs on LCM dosing, serum concentrations, and clearance in patients have produced mixed results.^14,19–24 Recognizing the dearth of LCM pediatric pharmacokinetic data, our objective was to evaluate the influence of age and concomitant ASM on the weight normalized dose-to-concentration ratio (DCR) and FDA LCM dosing recommendations in pediatric patients. A secondary objective was to identify those specific ASMs affecting LCM clearance and the influence of individual isozymes.

Materials and Methods

Data Source. This single-center, retrospective study was conducted at an accredited Level IV National Association of Epilepsy Center at Le Bonheur Children’s Hospital. Electronic medical records for our institution, which are housed in the Cerner database (North Kansas City, MO), were searched for patients who had a LCM serum concentration between October 2009 and June 2017. Data were extracted using Discern Analytics 2.0 (Cerner, North Kansas City, MO) and were exported into an Excel spreadsheet (Microsoft, Seattle, WA). All data were anonymized.

Data from hospitalized patients were included if the patient was over 1 month and ≤18 years of age and had received LCM prior to admission; and had at least 1 LCM serum concentration. Data were excluded if: 1) there were incomplete data; 2) LCM serum concentration was not detectable; 3) patients were outside the age requirements; and/or 4) renal or hepatic dysfunction were noted.

If a patient had ≥ 2 LCM concentrations, all samples were used provided the: 1) dose changed, 2) ASM regimen differed, 3) patient changed age groups, or 4) serum samples were collected more than a year apart. When more than 1 sample had the same dose and ASM regimen, the data were averaged, and the mean number was used. A sample was omitted if it was deemed to be an outlier, which we defined as a calculated DCR that was >20% of the averaged pool mean. If inclusion of a sample was unclear, 2 authors (SJP and MLC) agreed on inclusion or exclusion of that patient or data point. After these modifications, sera from a patient that appeared more than once were treated as independent events.

Study Population. Age, weight, sex, and race were recorded. Lacosamide dosage was normalized for body weight. Adjunctive ASMs were noted for each data point on a date that corresponded to the time of the LCM concentration. Doses and serum concentrations of the adjunctive ASM and other medications were not recorded.

Serum Concentrations. LCM serum concentrations were collected as clinically indicated. Although time of LCM serum collection was not recorded, ASM concentrations are generally collected before the morning dose. Collection was done without regard to fasting as food does not affect the rate or extent of absorption.¹ LCM concentrations were analyzed using a validated liquid-chromatography tandem mass spectrometry method by MedTox Laboratories (St. Paul, MN). This methodology demonstrated acceptable precision and accuracy over the tested concentration range (0.5–20 mg/L), with a lower limit of quantitation of 0.5 mg/L. Our center uses a target LCM serum concentration range of 2 to 12 mg/L.

Study Outcomes. DCR was used as a surrogate of clearance and was calculated as follows: DCR (mL/kg/hr) = Dose (mg/kg/day)/24 hr/day ÷ concentration (mg/L)/1000 mL/L. The most recent FDA recommendations were used in evaluating LCM dosing.¹ Patients were stratified into 1 of 4 age backets: 1 month to 2 years; ≥ 2 years to 6 years; ≥ 6 years to 12 years; and ≥ 12 years to ≤18 years. Activity of each adjunct ASM at the CYP2C9, CYP2C19, and/or CYP3A4 isozymes was stratified based on product information and published literature (Supplemental Table S1).^25–30 The degree of interaction (i.e., strong, moderate, weak, or neutral) was classified using primary and tertiary literature and all authors agreed with the assigned designation.

To examine the effect of concurrent ASM on LCM DCR, data were divided into groups based on the potential for a drug-drug interaction. Because there was no difference (p = 1) in DCR for LCM monotherapy [0.043 (0.030–0.0620); n = 98] vs LCM plus a neutral ASM [[0.043 (0.030–0.061); n = 151], these data were merged to form a single group. Some ASM have both inducing and inhibiting properties (e.g., felbamate, eslicarbazepine, oxcarbazepine, topiramate) and are considered as having a “mixed effect.” There was no difference (p = 0.26) in DCR between regimens containing LCM plus inducers and inhibitors [0.074 (0.045–0.011); n = 94]; vs those containing LCM plus a single ASM with “mixed” properties [0.060 (0.035–0.085); n = 76]; hence, these data were combined to form a single group. This analysis produced 4 distinct groups for assessment of LCM-ASM interactions: LCM plus neutral ASM; LCM plus ≥ 1 inducing ASM; LCM plus ≥ 1 inhibiting ASM; and regimens containing LCM plus both inducing and inhibiting ASM. When sample size was sufficient, DCR was also determined for individual ASMs and for the 3 specific isozymes responsible for LCM metabolism.

Statistical Analysis. Binary data were reported as count and percentage, and continuous variables were summarized as median (IQR). χ² and Fisher exact tests were used to examine the differences between categorical variables. Regression analysis was used to examine the correlations between LCM dose and LCM serum concentration and was further stratified by the FDA defined dosing weight groups. Regression analysis also compared the relationship between dose and serum concentration among patients receiving neutral ASM and those receiving inducing ASM. The relationship between DCR across the predefined age groups, as well as the FDA defined dosing weight groups and the impact of adjunctive ASM therapy was examined using Kruskal-Wallis test and a pairwise comparison of groups adjusted by the Bonferroni correction to account for multiple tests. A p value ≤ 0.05 was considered statistically significant. All analyses were conducted using IBM SPSS Statistics version 28.0 (Microsoft, Seattle, WA).

Results

A total of 1405 orders for LCM were identified in 529 patients with 646 sera points from 380 unique patients. Reasons for exclusion are noted in the (Supplemental Figure). Data points were stratified by: age (4 cohorts); FDA weight-based dosing recommendations (5 groups); and potential for ASM interactions with LCM (4 groups; Supplemental Figure). Characteristics for all data points are noted in Supplemental Table S2. The correlation between LCM dose and serum concentration was statistically significant (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

Impact of Age on DCR. Table 1 shows data stratified into 4 age brackets. The largest number of data points occurred in those ages ≥ 6 to 12 and ≥ 12 to ≤18 and the least were in those < 2 years. Although there was a statistical difference is sex among the 4 age groups, there was no statistical difference in DCR between males and females [0.053 (0.036–0.083)] vs [0.052 (0.035–0.079)], respectively (p = 0.46). For all data, DCR was significantly greater in white patients compared with black patients [0.054 (0.036–0.089)] vs [0.046 (0.030–0.072)], respectively (p = 0.01). Although the weight-normalized ­daily LCM dose was different across age groups, there was no statistical difference in serum LCM concentration (p = 0.217).

Table 1. Characteristics of Sera Stratified for Age Classifications (N = 646)

View Fullscreen

There was a significant difference in DCR between the age groups (Table 1). Those <2 years had the highest DCR and those ≥ 12 to ≤ 18 years had the lowest. As one moves across increasing age groups the DCR decreases by 30.7%, 50.5%, and 63.4%. DCR in those 1 month to < 2 year were statistically different from those ≥ 2 to 6 years, ≥ 6 to <12 years, and ≥ 12 to ≤ 18 years. The DCR in those ≥ 2 to < 6 years statistically differed from those ≥ 6 to <12 years and those ≥ 12 to ≤ 18 years. There was no difference between those ≥ 6 to <12 and ≥ 12 to ≤ 18 years of age. There was a weak (R² = 0.073) but significant (p < 0.001) negative correlation between DCR and age suggesting that as age increases DCR decreases.

Those < 2 years of age received more adjunctive ASM than the other 3 age groups. None of the sera in those < 2 years received LCM alone, while 17.6%, 15.4%, and 18.4% of those ≥ 2 to 6, ≥ 6 to 12, and ≥ 12 to 18 years, respectively, received monotherapy. The regimens of those < 2 years also contained the greatest number of strong inducers (27.4%) compared with those ≥ 2 to 6 (8.8%), ≥ 6 to 12 (11%), and ≥ 12 to 18 (10.1%) years (p < 0.001). There were no differences in the frequency of regimens including strong inhibitors between age groups (p = 0.69). Dose for the entire data set was 8.36 (5.92–11.16) mg/kg/day. There was a significant difference in weight adjusted dose across the 4 age cohorts with those < 2 and those ≥ 2 to 6 years receiving larger doses than the older age brackets. However, there was no difference is LCM serum concentrations across the age groups.

FDA Dosing Recommendations. Table 2 depicts information for the 5 weight categories used by FDA for dosing recommendations and Figure 2 shows LCM dose vs serum concentration for each group. Overall, 50.2% of doses fell within the FDA weight recommendations; however, 40.4% exceed the suggested dosing range. More than half of those ≥ 6 to 11 kg and ≥ 30 to 50 kg received doses larger than the recommended range. No infants ≥ 6 to 11 kg received a small dose, but of those < 6 kg and ≥11 to 30 kg, 18.7% and 14.3% respectively received smaller doses than recommended.

View Figure

• Download as Powerpoint
• Download Figure

Table 2. Characteristics of Sera Stratified by FDA Dosing Weights

View Fullscreen

Most (81.3%) of the 646 serum LCM concentrations were in the suggested target serum range; however, 4.5% and 14.2% were sub- or supratherapeutic, ­respectively. Those with a subtherapeutic concentration had a significantly higher DCR [0.131 (0.103–0.186)] than those with a therapeutic [0.053 (0.038–0.081)] or supratherapeutic [0.035 (0.025–0.054)] concentration.

While the sample size was small (n = 16), 25% of those < 6 kg had subtherapeutic concentrations. Regardless of weight category, those who received doses smaller than that recommended (77%) had low but “therapeutic” serum concentrations [3.6 (2.6–4.2)]; however, 31% of those had serum concentrations < 2 mg/L. While 45% of those with a subtherapeutic concentration received an ASM that was a weak inducer, none received phenobarbital. For the entire population, those given doses larger than those recommended had therapeutic serum concentrations (72%), but many serum concentrations were in the mid to upper end of the therapeutic range [8.2 (6.0–10.1)]. Most of those given a larger dose received an inducer (64%) with 27% of those receiving phenobarbital.

Influence of Adjunctive ASM on DCR. Excluding LCM, 31 different ASMs were given (Supplemental Table 3). Levetiracetam, clobazam, zonisamide, topiramate, valproate, phenobarbital, and oxcarbazepine were the most frequently prescribed ASM. Very few regimens included phenytoin, cannabidiol, carbamazepine, or eslicarbazepine. Adjunctive therapy was used in 84.8% of samples. Most polytherapy involved 1 (34.5%) or 2 (29.9%) additional ASMs, but 8.2% consisted of ≥ 4 ASM with the maximum of 7 in a single regimen.

Sera were assigned to 1 of 4 groups based on the potential to interact with LCM (Table 3). The largest and fewest number of sera were found in the LCM plus neutral and LCM plus only inhibitors groups (Group 3), respectively. Although 61.3% of regimens contained an ASM known to impact LCM clearance, most of these were considered weak, with only 12.1% and 1.4% being strong inducers or inhibitors, respectively. The correlation between dose and serum concentration was different for regimens not associated with a LCM-drug interaction compared with one containing an ASM known to induce clearance (Figure 3). DCR was higher in regimens containing LCM plus an inducer of LCM clearance [0.061 (0.042–0.101)] vs LCM plus a neutral ASM [0.043 (0.030–0.062)], respectively (p < 0.001). LCM dose was also larger in the LCM plus an inducer [10 (7.6–14.8) vs 7.4 (4.9–9.2) mg/kg/day], (p < 0.001); however, there was no difference in serum concentrations between the 2 groups.

View Figure

• Download as Powerpoint
• Download Figure

Table 3. Characteristics of Data Stratified by Drug-Drug Interaction Potential

View Fullscreen

There were no difference between LCM plus an inducer and LCM plus an inducer and/or inhibitors (p = 1) and the addition of an inhibitor alone did not significantly change clearance. There was also a significant difference in DCR for those given a single inducing ASM [0.056 (0.039–0.085)] and those given multiple inducers [0.085 (0.055–0.144)] (p < 0.001). Likewise, a regimen with a single strong inducer [0.091 (0.066–0.169)] was different from one containing multiple strong inducers [0.148 (0.130–0.161)], p < 0.05.

Table 4 compares DCR for regimens containing LCM as monotherapy and specific ASM. ASMs associated with the highest DCR were phenobarbital, topiramate, and clobazam. Phenobarbital had the greatest affect increasing the DCR by 2.6-fold. While DCR increased by 32.6% with clobazam and 72.1% with topiramate. There was no effect of lamotrigine, levetiracetam, zonisamide, oxcarbazepine, or valproate.

Table 4. Influence of Select ASM on Dose to Concentration Ratio When Compared With LCM Monotherapy

View Fullscreen

There were insufficient sample numbers to stratify DCR by strong and weak inducer of CYP2C9 and CYP2C19. Likewise, while cannabidiol, eslicarbazepine, and felbamate are strong or moderate inhibitors, their influence could not be evaluated due to either a small sample size or coadministration with other ASM that were strong inducers; hence, the inhibitor data represent only weak inhibitors, primarily valproate. The DCR for inducers of CYP2C9 [n = 19; 0.119 (0.075–0.168)] and CYP3A4 [n = 293; 0.067 (0.042–0.110)] were both greater (p < 0.001) than that for LCM monotherapy. Strong inducers of CYP3A4 significantly (p < 0.001) increased DCR [n = 70; 0.110 (0.075–0.160)]. Conversely, weak inducers did not significantly (p = 1) impact DCR [n = 223; 0.053 (0.038–0.813)]. Likewise, inhibitors of CYP2C9 [n = 58; 0.045 (0.031–0.063)], CYP2C19 [n = 137; 0.051 (0.035–0.081)], or CYP3A4 [n=56; 0.045 (0.032–0.065)] did not significantly affect DCR (p = 1, 0.24, and 1, respectively).

Discussion

Numerous papers have evaluated the efficacy and safety of LCM,^{4,14,19–23,31–37} but most focused on the adult population. Collectively the pediatric papers included 1241 sera, but only 4 stratified patients by age or weight (n = 552 sera).^14,19,21,22 Few patients were 6 months to < 2 years of age (n = 29) or between 2 and 6 years (n = 96), while most were ≥ 6 years with almost half being ≥12 year of age. By comparison, we report 646 sera from 380 patients. Although most sera in our study were from those ≥ 6 years (69%), we had more data points in those < 2 (n = 73) years of age and ≥ 2 to 6 (n = 125) than other papers combined.

Effect of Age on DCR. May et al¹⁴ noted that age significantly impacted LCM dosage and serum concentration in those < 12 years of age, but when estimating equation models adjusted for body surface area the effect of age was no longer significant. Svendsen et al¹⁸ also reported no difference in clearance between those ≤18 and 18 to 65 years of age; however, only 22 children were included. Winkler et al²² applied population pharmacokinetic modeling of pediatric data and found that age did not affect LCM clearance. Conversely, Larsen et al¹⁹ reported significant differences in LCM clearance with older children having lower clearances. Consistent with Larsen et al,¹⁹ we found that DCR decreased as age increased. Our finding may be due to the larger number of sera from those ≤ 6 year, the use of strong inducers in those ≤ 2 years, and/or the result of age-related physiologic changes that occur across childhood. Regardless, the negative correlation between age and DCR was weak and only accounted for about 6% of the DCR variability.

Consistent with product information,¹ and other reports,^12,17,18,21 we did not find that sex had a significant influence on the DCR. Product information notes no clinically relevant differences in LCM pharmacokinetics between races¹; however, we found that white patients had significantly higher DCR than black patients. While interesting, this maybe an artifact related to an uneven distribution in age or adjunctive ASM among race groups. Although, Zutshi et al³⁸ reported that black adults had lower serum LCM concentrations than white adults, this was based on clinically insignificant difference in serum concentrations (6.8 mg/L vs 7.1 mg/L). Unlike Zutshi et al,³⁸ we found no difference in serum concentrations between the 2 races.

FDA Dosing Recommendations. Similar to Rastogi and Ng⁶ and Contin et al,¹⁷ we found that LCM doses were larger than those recommended by FDA. Doses ranged from 4.7 to 8.4 mg/kg/day^{4,21–23,31–37} in other reports compared to our study where mean (10 ± 7.33) and median [8.36 (5.92–11.16)] mg/kg/day doses were much larger. Only 6 studies looked specifically at LCM dosing by weight brackets.^19–24 In prospective studies, Farkas et al,²⁰ and Ferreira et al²¹ recommended the upper limits of FDA dosing for those < 30 kg (12 mg/kg/day), 30 to 50 kg (8 mg/kg/day), and > 50 kg (400 mg/day). Likewise, Winkler et al²² recommended that the same doses would maintain serum concentration in the same range as adults who received 400 mg/day. Ishikawa et al²³ recommended that those ≤ 25 kg receive a maximum of 15 mg/kg/day (up to 400 mg/day) and those > 25 kg should receive 400 mg/day. Larsen et al¹⁹ simply noted that those < 50 kg receive larger doses than those ≥ 50 kg (6.4 vs 4.8 mg/kg/day). Lukka et al,²⁴ used pharmacometric modeling in those given LCM plus 1 ASM and recommended that in order to match the exposure seen in children ≥4 years, those ≥3 years needed the same dose as recommended by FDA for children ≥4 years (12 mg/kg/day), while children 1 to 3 years may need 13 to 14 mg/kg/day and infants between 1 month and 1 year may need 15 to 18 mg/kg/day (based on their actual age). Although some studies^20,21 noted adjunctive ASM, they did not assess their impact on LCM clearance or dose.

We also found that the FDA recommended dosing was inadequate as more than 30% of sera in each weight bracket required larger doses. This was especially true in those ≥ 6 to 11 kg and those 30 to 50 kg where more than half required > 12 mg/kg/day and 400 mg/day, respectively. Patients < 2 years also required significantly more LCM. The requirement for larger doses was especially true in those receiving any combination of inducers, but especially in those given strong inducers.

Few studies reported serum concentration data. Like Svendsen et al¹⁸ and Ishikawa et al,²³ we found a strong linear relationship between LCM dose and serum concentration. While most patients in their studies had mean or median values within the reference range of 2 to 12 mg/L, they noted that many were in the lower end of the reference range. May et al¹⁴ simply noted that concentrations were significantly lower in those < 6 (38%) and 6 to 12 (21%) years compared with adults. Although most of our data points had serum concentration within the reference range, we found a significant difference in serum concentrations across the 5 FDA weight brackets. Although the number of samples were small, a quarter of those < 6 kg had subtherapeutic values. This could be due to the larger number of regimens (94%) that contained an inducer.

Effect of Concomitant ASM. Few data points in our study involved LCM monotherapy (n = 98), which is consistent with other pediatric reports. It is not surprising that most patients received 1 or 2 additional ASM as pediatric patients may have inadequate seizure control on 1 ASM. This is especially true of those with various seizure syndromes and those requiring larger doses. Because we did not collect information on the specific epilepsy, we were unable to assess the impact of disease on number or type of ASM and LCM dose used.

Consistent with our report (Supplemental Table S3), levetiracetam was also the most used adjunctive ASM in other studies.^{8,13,18–21,34,35} ASM commonly used in our population were also noted by others and included topiramate,^{13,18–21,34} valproate,^{8,13,18–21,35} clobazam,^19,34 and lamotrigine.^19,21,34 Only Farkas et al,²⁰ Sanmartí-Vilaplana and Díaz-Gómez,³⁵ and our report noted the use of strong enzyme inducers (e.g., carbamazepine, phenobarbital, phenytoin). Farkas et al²⁰ and Sanmartí-Vilaplana and Díaz-Gómez³⁵ noted the use of carbamazepine in 29.2% and 56.1% of patients, respectively. Although our patients received carbamazepine, phenobarbital, phenytoin, there were insufficient numbers to assess the impact of carbamazepine and phenytoin as single agents.

Most studies, including ours, reported a significant impact of strong inducing ASM on LCM clearance and/or serum concentrations. Three studies noted that strong inducers carbamazepine,^15–17 phenytoin,^15–17 and phenobarbital^15,17 decreased LCM serum concentration by 30%^15,16,18 or increased dosage by 30%.¹⁷ In another predominately adult study, May et al¹⁴ found that carbamazepine, phenytoin, and phenobarbital decreased LCM serum concentrations by 30%, 32%, and 39%, respectively. Likewise, Heyman et al⁵ reported that carbamazepine and phenytoin reduce serum LCM concentrations by 30% to 40% and 25%, respectively. We found that pediatric patients given LCM plus an inducer had significantly higher DCR, with phenobarbital having the greatest affect. An unexpected finding in our study was the increase in LCM clearance that was associated with topiramate (mixed, weak affect ASM) and clobazam (weak affect ASM).

We also found that regimens containing more than 1 inducer had a higher DCR than those containing a single inducer. Likewise, the combination of 2 weak or 2 strong inducing ASM resulted in a significantly higher DCR than that noted with a single weak or strong ASM. Although Yamamoto et al¹⁶ concluded there was no synergistic effects in those receiving multiple inducers, our finding do suggest a synergistic effect, which should be researched further.

In our study, valproate, a weak inhibitor, was the only inhibiting ASM given in sufficient numbers to allow assessment. Our finding, that ASM known to inhibits of CYP2C9, 2C19, and 3A4 had no effect on DCR, are consistent with other reports.^{13,14,18,19,22} These finding are counter to product information, which notes that dose reduction may be necessary in patients with hepatic impairment who are taking strong inhibitors of CYP3A4 and CYP2C9.¹

Some ASM are both weak inducers of CYP3A4 and weak inhibitors of CYP2C19. Except for topiramate, we did not find a difference in the DCR for LCM ­monotherapy and LCM plus these ASM, suggesting that perhaps the inducing activity is either inconsequential or suspended by the inhibiting effects. Failure to inhibit LCM clearance is not surprising as omeprazole, a potent inhibitor of CYP2C19, did not produce clinically significant changes in LCM area under the concentration curve.³⁹ Likewise, genetic polymorphisms of CYP2C19 do not produce clinically significant changes in LCM pharmacokinetics.⁴⁰

Although phenytoin is a strong inducer of CYP2C9 and CYP2C19, it was only given to 6 patients; hence, it was not possible to evaluate its impact on LCM clearance via CYP2C19. Carbamazepine, phenobarbital, and phenytoin are strong inducers of CYP3A4 and carbamazepine and phenobarbital are weak inducers of CYP2C9. Although there was a significant increase in LCM clearance for drugs acting at both CYP2C9 and CYP3A4, it is likely that the impact on DCR occurs predominately via CYP3A4. ASM that are weak inducers of CYP3A4 and weak inhibitors of CYP2C9, CYP2C19, and CYP3A4 had little to no effect on LCM clearance.

Limitations. The main limitation of our study was its retrospective nature and reliance on clinical records. This is not unusual as 72% of papers investigating this topic were also retrospective. Because we did not collect information on the specific type of epilepsy, we were unable to assess the impact of disease on the number and type of ASM. Use of LCM for an epilepsy syndrome where LCM may not be the most effective drug could easily result in larger dosing due to poor patient response. We also were unable to investigate the impact of prematurity on DCR in the youngest age group as we did not record gestational age. That said, polymorphism of CYP2C19 do not impact LCM clearance; hence, maturation of this specific isozyme would not be a factor. Variability in the time of serum concentration collection relative to dosing may have had some influence on our results. Some patients may have also received other non-ASM that might have affected LCM clearance; however, these were not recorded. The number of regimens containing strong (i.e., cannabidiol) or moderate (eslicarbazepine, felbamate) inhibitors of LCM clearance was limited and may have influenced our findings. We also did not assess efficacy or toxicity in relation to dose or serum concentration; hence, we are not able to comment on the clinical application of our findings. Despite these limitations, our findings contribute to a better understanding of the use of LCM in the treatment of epilepsy in pediatric patients.

Conclusions and Recommendations

Although there was a negative correlation between age and DCR it was weak and only accounted for about 6% of the DCR variability. Administration of a strong inducer significantly increased LCM clearance and may necessitate a 30% increase in dosage. There was no effect of weak inhibitors on LCM clearance. Our analysis suggests some patients may require larger initial doses of LCM. We recommend those < 6 kg be given 14 to 16 mg/kg/day, those 6 to 11 kg receive 13 to 14 mg/kg/day and those 30 to 50 kg be given 12 mg/kg/day (up to 400 mg). We did not evaluate efficacy or toxicity; hence, these suggestions simply reflect those doses necessary to achieve a similar LCM exposure to that noted in our patient population. Regardless of dose, LCM serum concentration monitoring should be combined with clinical response to individualize dosing, especially in those receiving strong inducers. Additional prospective research should assess these dosing recommendations. Likewise, the impact of strong inhibitors on LCM clearance and the effects of synergy with multiple inducers should be investigated.