Time–Course Changes in Lamotrigine Concentration after Addition of Valproate and the Safety and Long-Term Tolerability of Lamotrigine–Valproate Combination Therapy

Abstract

The aim of this study was to evaluate the time–course changes in lamotrigine (LTG) concentration after addition of valproate (VPA) and the safety and tolerability of the combination therapy. We reviewed our therapeutic drug monitoring (TDM) database and found 345 patients on LTG who received add-on therapy with VPA. VPA had been added at least 12 weeks after patients finished stepwise LTG titration. Also, we retrospectively evaluated the LTG concentration after addition of VPA and the safety and long-term tolerability of LTG–VPA combination therapy. Plasma LTG concentration increased more than 1.5-fold within 15 d of addition of VPA and reached a peak at 30 d. The rate of increase in LTG concentration occurred in a VPA concentration-dependent manner. During the first 120 d after addition of VPA, adverse events were reported by 58 patients (16.8%), but no patient developed cutaneous reactions. Kaplan–Meier analysis showed estimated retention rates for LTG–VPA combination therapy of 74.5% at 5 years. At 5 years, the mean concentration of LTG was 11.1 µg/mL (43.3 µmol/L). Because addition of VPA leads to a marked increase in LTG concentration over a short period, TDM for LTG should be performed at the earliest from 14 d after starting VPA. At 120 d after starting VPA therapy, the higher LTG concentration due to addition of VPA is not associated with an increased risk of cutaneous reactions. Although LTG–VPA combination therapy increases LTG concentration, it is well tolerated and has a high long-term retention rate.

INTRODUCTION

Lamotrigine (LTG) is a widely prescribed antiseizure medication (ASM) that is used in monotherapy and as an adjunctive treatment for focal, focal to bilateral, and generalized tonic-clonic seizures. It is metabolized by the uridine-5′-diphosphate-glucuronosyl-transferases (UGTs), such as UGT1A4 and UGT2B7, and UGT inducers or inhibitors can have large effects on its concentration. For example, concomitant use of enzyme-inducing ASMs (inducers), such as phenytoin (PHT), carbamazepine (CBZ), and phenobarbital (PB), reduces the concentration of LTG by 40 to 70%. In contrast, valproate (VPA) is known to be an inhibitor of UGTs and to increase the LTG concentration by 50 to 150%.^1–3) Thus, therapeutic drug monitoring (TDM) is clinically useful for assessing LTG concentration and guiding dosage adjustments.

According to the TDM guidelines for ASMs published by the International League Against Epilepsy in 2008, the reference range for the serum concentration of LTG is 2.5 to 15 µg/mL.¹⁾ Although the optimal effective concentration of LTG is unclear, clinical response to LTG occurs in a concentration-dependent manner. The risk of adverse events increases with higher LTG concentrations, and the incidence at concentrations of 5.0 to 10, 10 to 15, and greater than 20 µg/mL was found to be 14, 24, and 59%, respectively.⁴⁾ LTG should be started at a low dose and titrated slowly because a high blood concentration during the initial stage is associated with an increased risk of serious cutaneous reactions, such as Stevens–Johnson syndrome and toxic epidermal necrolysis.⁵⁾

Generally, 60 to 70% of epilepsy patients respond to standard ASMs, but 30% of patients fail to achieve seizure control and develop refractory epilepsy.⁶⁾ These patients are generally treated with multiple ASMs, and their treatment regimen often has to be changed, e.g., by adding or discontinuing drugs.

According to an in vitro animal model study by Taing et al., the combination of LTG and VPA has a synergistic effect.⁷⁾ Brodie and Yuen enrolled 357 epilepsy patients who were not fully controlled with ASM monotherapy and found that adding LTG to VPA produced a significantly better response than adding it to CBZ or PHT.⁸⁾ Thus, the pharmacokinetic and pharmacodynamic interactions between LTG and VPA can lead to anticonvulsant synergy.^8–10)

To date, few studies have examined the pharmacokinetic effect of adding UGT inhibitors to LTG treatment. There is also limited information on the safety and tolerability of adding VPA to LTG. Therefore, the aim of this study was to evaluate the time course of LTG concentration after addition of VPA and the safety and tolerability of the combination treatment.

MATERIALS AND METHODS

Participants

Between September 2009 and December 2015, we performed routine TDM for LTG and obtained 23823 plasma samples from 2722 epilepsy patients. In the present study, we retrospectively reviewed the TDM database and found 345 patients who had received LTG and add-on therapy with VPA. We retrospectively evaluated the LTG concentration after addition of VPA and the safety and long-term tolerability of LTG–VPA combination therapy by reviewing the patients’ clinical records from September 2009 to December 2021. This study did not include pregnant women or women on oral contraceptives because these conditions can affect LTG pharmacokinetics. PHT, CBZ, and PB (including primidone) were defined as inducers.

The study was approved by the institutional ethics committee of the National Epilepsy Center (Shizuoka, Japan). The ethics committee waived the requirement for informed consent because of the retrospective nature of the study (opt-out agreements).

Measurement of LTG Concentration

To avoid cutaneous reactions, VPA was added to LTG at least 12 weeks after the start of LTG therapy. This study included only patients in whom the dose of LTG and inducers had not been changed for at least 28 d. The date of the addition of VPA to LTG was defined as day 0, and baseline was defined as 24 weeks before the start of VPA add-on therapy. If two or more plasma samples were obtained from a single patient at baseline, the analyses used the mean value of the samples collected while the patient was receiving the maintenance dose of LTG. Plasma concentrations of LTG were assayed by the method reported previously.¹¹⁾ In this study, we evaluated plasma samples collected 3 to 5 h after administration of LTG (non-trough sampling); samples were collected from each patient at same time points (within ±1 h). The conversion factor for LTG was as follows: 1 µg/mL = 3.9 µmol/L.

Additionally, we evaluated the change in LTG concentration during the first 120 d after addition of VPA. To do so, we divided the 120 d into 3 observation periods: days 1 to 15, 16 to 30, and 31 to 120. In each period, we calculated (a) the concentration-to-dose ratio (CD ratio) by dividing the LTG concentration in µg/mL by the dose adjusted for body weight in mg/kg and (b) the change ratios by dividing the LTG CD ratio at each time point by the baseline CD ratio. We also evaluated when the LTG concentrations started to increase.

Safety and Tolerability of LTG after Addition of VPA

The adverse events associated with LTG–VPA combination therapy were retrospectively assessed by reviewing the clinical records. We defined adverse events due to combination therapy as a reduction of LTG dose or discontinuation of LTG or VPA or both. The observation period was set at 120 d from commencement of VPA add-on therapy. Additionally, to evaluate the long-term tolerability of the combination therapy, we retrospectively reviewed the clinical records of patients from September 2009 to December 2021. The estimated retention rate for combination therapy was calculated by the Kaplan–Meier method, and the observation period was set at 5 years (1825 d).

Statistical Analysis

The group differences in change ratios were evaluated by analysis of covariance (ANCOVA) or ANOVA. When the ANOVA result was significant, the post hoc Games/Howell test was applied. Also, ANCOVA was performed with covariates such as age, sex, baseline CD ratio, or VPA concentration and Bonferroni post hoc tests. Results are expressed as means or odds ratios with 95% confidence intervals (CI). Statistical analyses were performed with SPSS software Ver 25.0 (IBM Corp., Armonk, NY, U.S.A.).

RESULTS

Baseline Characteristics

The baseline characteristics of the 345 patients are shown in Table 1. All patients were unresponsive to standard ASM regimens and had inadequate control of seizures after starting LTG therapy. Our study included 61 pediatric patients (aged 1 to 16 years, 17.7%) but only 5 older adult patients aged 65 to 69 years. A total of 223 patients (64.6%) were treated with inducer regimens, and 56 patients received multiple inducers. No patients were taking oxcarbazepine or eslicarbazepine because these ASMs are not approved in Japan. The majority of patients (52.2%) continued LTG therapy for more than 1 year.

Table 1. Baseline Characteristics of 345 Patients with Epilepsy

	Mean (95% CI) or number of patients
Age, year	31.6 (30.0–33.2)
Sex, male/female	181/164
Body weight, kg	55.0 (53.0–56.9)
Lamotrigine dose, mg/d	315.5 (301.5–329.4)
Lamotrigine dose, mg/kg/d	5.9 (5.7–6.2)
Lamotrigine concentration, µg/mL	4.9 (4.6–5.2)
Duration of lamotrigine treatment
≥ 6 months, <1 year	94
≥1 year, < 2 years	182
< 2 years	69
Concomitant ASMs
Enzyme-inducing ASMs	223
Phenytoin	117
Phenobarbital	58
Carbamazepine	106
Zonisamide	56
Benzodiazepines	94
Levetiracetam	54

ASMs, antiseizure medications; CI, confidence interval. Lamotrigine concentration (µmol/L) = 3.9 × (µg/mL).

Time Course of LTG CD Ratio after Addition of VPA

Of the 345 patients, 49 underwent measurement of LTG concentration on days 1 to 15, 16 to 30, and 31 to 120. Figure 1 shows the time course of change in the LTG CD ratio after coadministration of VPA. At all observation points, mean change ratios were significantly higher in patients without inducers than in those with inducers (Bonferroni test, p < 0.001). In the inducer group, the mean VPA concentrations at days 1 to 15, 16 to 30, and 31 to 120 were 16.8, 27.5, and 27.6 µg/mL, respectively. In contrast, patients who were not using inducers tended to have a higher VPA concentration at the 3 measurements (24.9, 37.7, and 46.0 µg/mL, respectively). Because the VPA concentration differed between the inducer and non-inducer groups, we calculated the adjusted mean change ratio (Supplementary Table S1). As presented in the Table, the results adjusted for VPA concentration indicated a similar pattern between the two groups. Regardless of whether inducers were used or not, the CD ratio of LTG was higher during days 1 to 15. Moreover, the CD ratio reached a peak at 30 d after addition of VPA.

Fig. 1. Time–Course Changes in the Concentration-to-Dose Ratios of Lamotrigine

The figure shows the results from 49 patients in whom the lamotrigine concentration was measured at each time point. Open circles, patients with inducers (n = 16); closed circles, patients without inducers (n = 33). Analysis of covariance adjusted for age, sex, and baseline concentration-to-dose ratio of lamotrigine: * p < 0.001 vs. inducer group. CD ratio, concentration-to-dose ratio.

Table 2 shows the change in the CD ratio of LTG between days 1 and 15. TDM for LTG was performed in 120 patients during the first 15 d, which were subdivided into the following three periods: days 1 to 5, 6 to 10, and 11 to 15. Because the VPA dose and concentration differed between these 3 groups, change ratios were adjusted by the covariates. LTG concentration started to increase between days 1 and 5, and the mean CD ratio was 1.47-fold higher on day 5 than at baseline. After day 6, the LTG CD ratio tended to increase, but there was no significant difference between the 3 groups.

Table 2. Comparison of Mean Concentration-to-Dose Ratio in 120 Patients Treated with Lamotrigine and Add-on Valproate

	Time after starting VPA			p-Value
	1–5 d	6–10 d	11–15 d
Number of patients, n	27	57	36
LTG dose, mean (95% CI), mg/kg	6.2 (5.1–7.3)	5.9 (5.2–6.5)	6.2 (5.5–6.9)	NS
CD ratio of LTG, mean (95% CI), µg/mL/mg/kg	1.24 (1.05–1.45)	1.42 (1.24–1.60)	1.36 (1.13–1.60)	NS
VPA dose (95% CI), mg/kg	3.3 (2.5–3.9)	4.7 (3.9–5.6)	5.8 (4.7–6.6)	< 0.01
VPA concentration (95% CI), µg/mL	15.4 (12.0–18.8)	22.7 (18.7–26.7)	26.5 (22.0–31.1)	< 0.005
Change ratio
Crude value, mean (95% CI)	1.34 (1.19–1.49)	1.66 (1.50–1.82) †	1.72 (1.58–1.86) †	< 0.01
Adjusted value,* mean (95% CI)	1.46 (1.28–1.63)	1.65 (1.53–1.77)	1.65 (1.49–1.80)	NS

The table shows the results of 120 patients who underwent therapeutic drug monitoring for lamotrigine within 15 d after addition of valproate. CD ratio, concentration-to-dose ratio; CI, confidence interval; LTG, lamotrigine, 1 µg/mL = 3.9 µmol/L; VPA, valproate, 1 µg/mL = 6.93 µmol/L. *Analysis of covariance adjusted for age, sex, and valproate concentration. ^†p < 0.05 versus days 1 to 5, post hoc Games/Howell test.

When the relation between the CD ratio of LTG and the VPA concentration was investigated in 120 patients, a weak positive correlation was found (Spearman’s analysis; r = 0.30, p < 0.005).

Effects of VPA Concentration on the CD Ratio of LTG

The patients were divided into 3 subgroups on the basis of VPA concentrations (<20; ≥20, <40; and ≥40 µg/mL). The time course of the difference in the change ratio between these 3 subgroups is shown in Fig. 2. Despite the low VPA concentration (<20 µg/mL), the LTG CD ratio increased 1.3-fold over days 1 to 15. In all 3 of the VPA concentration subgroups, the CD ratio increased significantly as the VPA concentration increased (ANOVA, p < 0.001).

Fig. 2. Influence of Valproate Concentration on the Change Ratio

There were significant differences between the 3 valproate concentration groups (<20; ≥20, <40; and ≥40 µg/mL) at each time point (days 1–15, 16–30, and 31–90; ANOVA, p < 0.001). * p < 0.001, Games/Howell post hoc test. CD ratio, concentration-to-dose ratio.

Effects of Age on the CD Ratio of LTG

Supplementary Table S2 compares the change ratio of LTG between adult and pediatric patients (see the supplementary materials). Although there were no significant differences between adult and pediatric patients, adult patients tended to have a higher change ratio.

Safety and Tolerability after Addition of VPA

During the first 4 months, 75 of the 345 patients (21.7%) underwent a dose reduction of LTG; the mean concentration in these patients was 10.7 µg/mL. However, adverse events, i.e., events leading to discontinuation or dose reduction of VPA or LTG, occurred in 58 patients (16.8%). The most common adverse events were dizziness (n = 25), diplopia (n = 21), anorexia/vomiting (n = 17), somnolence (n = 8), headache (n = 6), psychological symptoms (n = 5), seizure worsening (n = 5), and insomnia (n = 3). During the observation period, 1 case of LTG-associated pure red cell aplasia was found. Although none of the patients developed cutaneous adverse reactions during the first 4 months, 1 patient discontinued both VPA and LTG due to a skin rash 308 d after addition of VPA. Also, 1 patient died of brainstem hemorrhage 1511 d after commencing combination therapy.

The Kaplan–Meier analysis showed estimated retention rates for LTG–VPA combination therapy of 85.9% at 1 year, 78.9% at 3 years, and 74.5% at 5 years (Fig. 3). The mean concentrations of LTG at 3 and 5 years were 11.9 µg/mL (n = 229; 95% CI, 11.1–12.6) and 11.1 µg/mL (n = 191; 95% CI, 10.4–11.9), respectively.

Fig. 3. Long-Term Retention Rate of Lamotrigine and Valproate Combination Therapy

DISCUSSION

In this study, we elucidated the time course of changes in LTG concentration after addition of VPA in patients with epilepsy. During the first 5 d, an inhibitory effect of VPA was observed. On day 14 after addition of VPA, the LTG concentration reached a steady state and was 1.5- to 2-fold higher than before VPA treatment. Therefore, patients must be monitored carefully for adverse events as soon as they start add-on treatment with VPA, and TDM for LTG is recommended after day 14.

Regardless of the time course of concentrations, even low VPA concentrations (<20 µg/mL) showed a marked inhibitory effect on LTG metabolism, and the inhibition effect was enhanced in a VPA concentration-dependent manner. According to a pharmacokinetic study in 6 healthy volunteers, LTG total clearance decreased by 21% from 2 d after adding 200 mg of VPA.^12,13) Gidal et al.¹⁴⁾ found that adding VPA reduced LTG clearance at a very low dose (125 mg/d) of VPA in healthy individuals. Also, a population pharmacokinetic model developed by Xu et al.¹⁵⁾ showed that LTG clearance was significantly influenced by body weight and VPA concentration (and decreased linearly). These studies support our findings. Accordingly, when physicians add VPA to LTG, they must monitor the concentration not only of LTG but also of VPA.

Generally, concomitant use of VPA is associated with increased risk of cutaneous adverse reactions. A Japanese phase IV study evaluated a practical clinical method for switching women with epilepsy from VPA monotherapy to LTG monotherapy.¹⁶⁾ In the study, LTG was first added to VPA monotherapy, then the VPA dose was tapered off. Despite the use of the standard titration as described in the package insert of LTG, 8 out of 33 patients (24.2%) developed LTG-related skin rash. In general, rapid elevation of LTG concentrations during the early treatment phase can increase the occurrence of LTG-related skin lesions and rash. Suzuki et al. reported that a plasma LTG concentration exceeding 4.4 µg/mL in week 2 is a risk factor for LTG-related rash.¹⁷⁾ In our study, although the LTG concentration increased by about 1.5- to 2.0-fold during the first 15 d, none of the patients developed an LTG-related rash. One patient did discontinue LTG and VPA due to a cutaneous adverse reaction, but the event occurred after 300 d of LTG–VPA combination therapy and the symptoms remained unchanged after discontinuation of LTG and VPA, so, we believe that the cutaneous reaction was not related to LTG. Therefore, although it is well known that the dose needs to be carefully titrated when starting LTG treatment to avoid cutaneous adverse reactions, our results indicate that once patients complete the dose-titration phase (≥12 weeks after starting LTG), a subsequent elevation of the LTG concentration may not be associated with an increased risk of cutaneous reactions. However, the incidence of severe cutaneous reactions is generally less than 0.1%, so further accumulation of cases is needed to confirm that this treatment approach is safe.

Generally, LTG has a higher retention rate than topiramate and levetiracetam (one study found a 2-year retention rate of 69.2% with lamotrigine, 45.8% with levetiracetam, and 38.3% with topiramate).¹⁸⁾ Furthermore, Wong et al. reported that patients receiving concurrent VPA were more likely to remain on long-term treatment with LTG.¹⁹⁾ In our analysis, LTG–VPA combination therapy showed good tolerability and a high long-term retention rate (74.5% at 5 years). Also, patients who were receiving long-term combination therapy had relatively high plasma LTG concentrations compared with the reference range of this drug. Although we could not evaluate the clinical response, higher LTG concentrations may be associated with a more favorable clinical response than lower concentrations.

Recently, the US Food and Drug Administration warned that clinically relevant concentrations of LTG could increase the risk of sudden death in patients with cardiac disease.²⁰⁾ Because it is unknown whether LTG–VPA combination therapy may increase the risk of sudden death, careful electrocardiogram monitoring should be performed, and future studies should assess long-term safety over at least 10 years.

Our study has several limitations. It was retrospective, and only 49 out of 345 patients (14.2%) had measurements of LTG concentration in all 3 time periods. Likewise, the titration schedule for VPA was different in each patient, so to calculate the LTG CD ratio, we had to adjust the mean change ratio by VPA concentration. Furthermore, although the plasma samples used to compare LTG CD ratios were obtained from each patient at the same time, we could not perform trough sampling. In this study, we defined an adverse event as an event leading to dose reduction or withdrawal of LTG or VPA. When other actions were taken (e.g., observation without dose reduction or dose reduction of other ASMs), the events were not classified as adverse events. Thus, our study may have underestimated the incidence of adverse events.

CONCLUSION

After addition of VPA to LTG, the LTG concentration increases about 1.5-fold over a short period. However, the elevation is not associated with an increased risk of adverse cutaneous reactions. Although LTG–VPA combination therapy leads to a higher LTG concentration, it has good tolerability and a high 5-year retention rate.

Acknowledgments

This study was funded by a JSPS KAKENHI Grant (Number: JP 23K06250).

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

This article contains supplementary materials.

REFERENCES

• 1) Patsalos PN, Berry DJ, Bourgeois BF, Cloyd JC, Glauser TA, Johannessen SI, Leppik IE, Tomson T, Perucca E. 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, 49, 1239–1276 (2008).
• 2) Patsalos PN, Spencer EP, Berry DJ. Therapeutic drug monitoring of antiepileptic drugs in epilepsy: a 2018 update. Ther. Drug Monit., 40, 526–548 (2018).
• 3) Johannessen Landmark C, Johannessen SI, Patsalos PN. Therapeutic drug monitoring of antiepileptic drugs: current status and future prospects. Expert Opin. Drug Metab. Toxicol., 16, 227–238 (2020).
• 4) Hirsch LJ, Weintraub D, Du Y, Buchsbaum R, Spencer HT, Hager M, Straka T, Bazil CW, Adams DJ, Resor SR Jr, Morrell MJ. Correlating lamotrigine serum concentrations with tolerability in patients with epilepsy. Neurology, 63, 1022–1026 (2004).
• 5) Wong IC, Mawer GE, Sander JW. Factors influencing the incidence of lamotrigine-related skin rash. Ann. Pharmacother., 33, 1037–1042 (1999).
• 6) Kwan P, Brodie MJ. Early identification of refractory epilepsy. N. Engl. J. Med., 342, 314–319 (2000).
• 7) Taing KD, O’Brien TJ, Williams DA, French CR. Anti-epileptic drug combination efficacy in an in vitro seizure model—phenytoin and valproate, Lamotrigine and Valproate. PLOS ONE, 12, e0169974 (2017).
• 8) Brodie MJ, Yuen AW. Lamotrigine substitution study: evidence for synergism with sodium valproate? Epilepsy Res., 26, 423–432 (1997).
• 9) Machado VH, Palmini A, Bastos FA, Rotert R. Long-term control of epileptic drop attacks with the combination of valproate, lamotrigine, and a benzodiazepine: a ’proof of concept,’ open label study. Epilepsia, 52, 1303–1310 (2011).
• 10) Thome-Souza S, Valente KD. Valproate and lamotrigine in pediatric patients with refractory epilepsy: after the first year. Pediatr. Neurol., 48, 436–442 (2013).
• 11) Yamamoto Y, Inoue Y, Matsuda K, Takahashi Y, Kagawa Y. Influence of concomitant antiepileptic drugs on plasma lamotrigine concentration in adult Japanese epilepsy patients. Biol. Pharm. Bull., 35, 487–493 (2012).
• 12) Yuen AW, Land G, Weatherley BC, Peck AW. Sodium valproate acutely inhibits lamotrigine metabolism. Br. J. Clin. Pharmacol., 33, 511–513 (1992).
• 13) Rambeck B, Wolf P. Lamotrigine clinical pharmacokinetics. Clin. Pharmacokinet., 25, 433–443 (1993).
• 14) Gidal BE, Anderson GD, Rutecki PR, Shaw R, Lanning A. Lack of an effect of valproate concentration on lamotrigine pharmacokinetics in developmentally disabled patients with epilepsy. Epilepsy Res., 42, 23–31 (2000).
• 15) Xu S, Liu L, Chen Y, Liu M, Lu T, Wang H, Liu S, Zhao M, Zhao L. Population pharmacokinetics of lamotrigine co-administered with valproic acid in Chinese epileptic children using nonlinear mixed effects modeling. Eur. J. Clin. Pharmacol., 74, 583–591 (2018).
• 16) Yamamoto T, Kubota Y, Murayama H, Ozeki H, Numachi Y, Ikeda A. Appropriate conversion from valproate monotherapy to lamotrigine monotherapy in Japanese women with epilepsy. Epilepsy & Seizure., 8, 21–31 (2016).
• 17) Suzuki T, Mihara K, Nagai G, Kagawa S, Nakamura A, Nemoto K, Kondo T. A high plasma lamotrigine concentration at week 2 as a risk factor for lamotrigine-related rash. Ther. Drug Monit., 42, 631–635 (2020).
• 18) Bootsma HP, Ricker L, Hekster YA, Hulsman J, Lambrechts D, Majoie M, Schellekens A, de Krom M, Aldenkamp AP. The impact of side effects on long-term retention in three new antiepileptic drugs. Seizure, 18, 327–331 (2009).
• 19) Wong IC, Mawer GE, Sander JW, Lhatoo SD. A pharmacoepidemiologic study of factors influencing the outcome of treatment with lamotrigine in chronic epilepsy. Epilepsia, 42, 1354–1358 (2001).
• 20) FDA Drug Safety Communication. “Studies show increased risk of heart rhythm problems with seizure and mental health medicine lamotrigine (Lamictal) in patients with heart disease (03 Mar, 2021).”: ‹https://www.fda.gov/media/147183/download›.