JLE

Epileptic Disorders

MENU

Efficacy of lamotrigine in idiopathic generalized epilepsy syndromes: a video-EEG-controlled, open study Volume 1, issue 3, Septembre 1999

Figures

See all figures

During the last few years many new antiepileptic drugs (AED) have been marketed to treat epilepsy refractory to classic AED. Nevertheless, clinical development is principally designed for licensing and not for clinical practice use. So a majority of trials are of little value in terms of treatment choice in a given patient [1]. Lamotrigine (LTG) is licensed in about 80 countries as add-on therapy in severe refractory epilepsy. Placebo-controlled, double-blind studies have shown efficacy in partial seizures with or without secondary generalization [2-7] and in Lennox-Gastaut syndrome [8, 9], whereas in one open study [10] it has been shown to aggravate severe myoclonic epilepsy of infancy. Open label studies in adults and in children suggested efficacy in different, generalized seizure types [11-13] and in cryptogenic and symptomatic generalized epilepsy syndromes, (for review see [14, 15]). Recently, this was confirmed in a short-term, double-blind, placebo-controlled study in 26 patients with treatment-resistant generalized epilepsy, which included 21 patients with IGE. However, differences in LTG response between individual syndromes were not examined [16].

Information on the clinical usefulness of LTG in specific syndromes of idiopathic generalized epilepsy (IGE) has been scarce until now. There are a few reports of partial efficacy in intractable childhood absence epilepsy [13, 17, 18] and in juvenile myoclonic epilepsy [19, 20].

We report the results of a prospective, open-label study in well-defined patients with IGE, in order to assess the LTG response profile as add-on or monotherapy in specific IGE syndromes. The primary outcome parameter was freedom from seizure, which was monitored by long-term video-EEG recordings, as the short and inconspicuous absence seizures are often not noticed by the patients and can be easily missed by parents or teachers [21].

Method

Patients

We studied 47 patients with IGE who had been followed at the Epilepsy Unit of the University Hospital of Strasbourg between 1993 and 1997. Patients were included in the study if their clinical history and long-term video-EEG recordings allowed diagnosis of unambiguous IGE syndrome according to the ILAE 1989 classification [22]. All patients had the usual neurological examination and 28 patients had neuroimaging (CT and/or MRI) performed to exclude structural brain lesions. LTG therapy was justified for the following reasons: seizures uncontrolled by previous conventional AED (22 patients), serious side effects on effective previous medication (12 patients), and persistent seizures with drug-related side effects (weight gain, alopecia, cognitive problems, tremor) for 13 patients. A summary of patient characteristics is given in table I. The data from a patient excluded early on from the study because of the appearance of a rash, are not included in the table.

Study design

In this study, patients received LTG initially as add-on therapy following a standard protocol. LTG monotherapy was progressively installed if a patient's seizures were controlled by this combination therapy but they still experienced adverse events from concomitant AED.

In all patients, video-EEG monitoring included at least a two-hour recording during wakefulness, including hyperventilation and intermittent photic stimulation. A whole-night sleep recording was performed in 38 patients, and in some patients EEG was carried out following sleep deprivation to facilitate seizures. In all patients with childhood (CAE) (n = 12) or juvenile absence epilepsy (JAE) (n = 12), seizures had been recorded prior to inclusion. In 8 of the 15 patients with juvenile myoclonic epilepsy (JME), myoclonias after awakening had been recorded, with additional recording of absences in 3 patients and a generalized tonic-clonic seizure (GTCS) in one patient. In patients with epilepsy with "grand-mal" seizures on awakening (EGMA) (n = 5), no seizures could be recorded due to their paucity, but generalized EEG interictal discharges were observed. In pure photosensitive epilepsy (PPE) (n = 3), myoclonias or GTCS could be induced by intermittent photic stimulation in all patients.

During the follow-up period of 25.5 ± 11.5 months (range 5-46 months), at least one awake and/or sleep video-EEG recording was performed in all patients with CAE or JAE and in all but four patients with other syndromes.

LTG was added progressively prior to AED medication. LTG doses started at 25 mg/day (adults) and 0.2 mg/kg body weight/day (children less than 12 years) for patients not receiving concomitant sodium valproate (VPA). Patients on VPA therapy received half the dose. LTG was increased by steps of 25 mg every fortnight. With this regimen, no allergic reaction occurred. One patient with JME, being treated with 2,000 mg VPA/d and in whom LTG was erroneously started at 100 mg/d, developed a cutaneous rash after 10 days; LTG had to be stopped: there was no attempt to reintroduce it at a smaller dose.

Evaluation of efficacy and tolerability

The primary efficacy measure was complete control of seizures at the end of the follow-up for at least five months, but most patients who responded to LTG were seizure-free throughout the whole follow-up period. Secondary efficacy measures included control of single seizure types, disappearance of EEG abnormalities during wakefulness, disappearance or reduction of photosensitivity and of discharges during hyperventilation.

Other outcome measures were i) the dose reduction of concomitant AED, resulting in a notable improvement in quality of life, especially in patients in whom LTG was introduced because of side effects of prior medication, and ii) the occurrence of adverse effects on LTG therapy.

Results

Seizure frequency

Table II summarizes the results of the primary efficacy measure, freedom from seizure, and the reduction of seizure frequency in all syndromes based on both the subjective anamnestic impression and the result of the video-EEG recordings performed after LTG introduction. An analysis of absence seizure subtypes across syndromes is given in table III.

In all patients that were not seizure-free at the end of the follow-up period, one of the four following features was present:

1) absences with a mild atonic component: both children with CAE who did not improve had this type of absence seizure;

2) a long-standing history of GTCS: one patient with EGMA had had GTCS for 52 years; however her GTCS was reduced from 8-12 to 1-2 GTCS/month. In a JAE patient with a 34-year history of GTCS and who usually had 1 GTCS/week before LTG introduction, complete control of GTCS was achieved with the persistence of rare absences;

3) the occurrence of absences with eyelid myoclonia and a history of rare GTCS was noted in one patient with CAE, in one patient with JAE and in one patient with JME;

4) poor compliance, observed in the EGMA patient classified as unchanged and in whom LTG was replaced by phenobarbital as he experienced repeated withdrawal GTCS.

Six patients with JME were classified as not seizure-free, although GTCS and absence seizures were completely controlled. They continued to have only myoclonias of the extremities after awakening and this, in most cases, with a notable reduction in frequency and strength.

EEG changes as secondary outcome variables

Tables IV, V and VI summarize the results for the additional outcome measures: disappearance of EEG abnormalities during wakefulness, disappearance or reduction of photosensitivity and of discharges during hyperventilation.

Adjustment of concomitant medication

VPA dosage could be reduced substantially in most patients. In CAE, the mean VPA dose could be reduced from 860 ± 590 to 530 ± 440 mg/d, in JAE from 1,220 ± 450 to 510 ± 240 mg/d, and in JME from 1,410 ± 440 to 1,070 ± 570 mg/d.

At the time of LTG introduction, 7 out of 12 patients with CAE were on bitherapy regimens combining VPA and ethosuximide (ESM). In all of them, ESM was able to be stopped after effective doses of LTG were reached. In all other syndromes taken together, a total of 10 patients were on different additional AED (carbamazepine (CBZ) n = 2, CBZ plus phenobarbital (PB) n = 1, phenytoin (PHT) n = 1, PB plus clonazepam n = 1, ESM plus PB plus clobazam (CLB) n = 1, CLB n = 2, primidone n = 1, oxcarbazepine n = 1) in combination with VPA. In all but one patient with EGMA, these additional AED were able to be stopped after LTG introduction. In this patient, CBZ was stopped, PB was reduced from 200 to 60 mg/d and CLB was reduced from 60 to 30 mg/d with complete seizure control.

At the end of the follow-up period, 42 patients were on either a bitherapy combining VPA and LTG (n = 30), on LTG monotherapy (n = 11) or in one patient with CAE, medication was able to be stopped because of natural remission. In two patients with EGMA and two patients with JAE, a combination of LTG with PB or PB plus CLB had to be employed to obtain complete seizure control.

Adverse effects and quality of life

No patient complained of adverse effects after LTG introduction. Only in one 5 year-old child with CAE did the parents report a transient tremor, agitation and a difficulty to concentrate after LTG introduction, which soon disappeared. Positive side effects possibly due to the reduction of the concomitant AED dosage, especially a weight reduction for patients on VPA was documented in 7 out of 18 patients with a weight loss ranging from 4 kg up to 16 kg (median 8 kg), resulting in a clearly improved quality of life in addition to the benefit due to seizure control.

Lamotrigine monotherapy

The 11 seizure-free patients on LTG monotherapy were distributed as follows: 3 patients with CAE, 4 patients with JAE, 2 patients with JME and 2 patients with PPE. In adult patients, the mean LTG dosage as monotherapy necessary to obtain complete seizure control was 280 ± 130 (range 100-400) mg/d, in children under 12 years of age it was 50 ± 25 (range 25-75) mg/d. The mean follow-up time for LTG monotherapy was 26 ± 13 (range 4-39) months. In 3 other patients with JAE, a LTG monotherapy trial failed, as absences reappeared after VPA had been stopped. The same was true for 2 patients with JME, in whom myoclonias reappeared after the discontinuation of VPA but they remained free of GTCS. After reintroduction of VPA, seizures were again completely controlled in all these patients.

Discussion

Our results suggest an overall high level of long-term efficacy and a good tolerability of lamotrigine in all IGE syndromes of childhood and adulthood examined. Nonetheless, a differential efficacy was observed in different syndromes and seizure types, as well as in single patients with clinical features across syndromatic borders. This leads to a discussion of difficulties encountered in antiepileptic drug trials using the current international classification of epileptic syndromes.

The highest response rates to LTG were observed in the absence epilepsies. In CAE, 9 out of 12 patients (75%) were seizure-free at the end of the follow-up period. In JAE, 10 out of 12 patients (83%) became seizure-free. These response rates are comparable to that reported by Ferrie et al. [12], of 64% seizure-free patients with typical absence seizures in a study cohort of 15 patients with different IGE syndromes receiving LTG as add-on treatment. Farrell et al. [18] reported 4 children with CAE, 3 of whom improved, with a seizure reduction of > 50% with add-on LTG. In another study, 3 out of 9 children with refractory absence epilepsy became seizure-free, and 4 children improved > 50% [17]. Bélanger et al. [13] described 5 children with refractory idiopathic epilepsies with absences, all of whom improved on adjunctive LTG, with one child becoming seizure-free.

In JME, complete control of GTCS was achieved in 13 out of 14 patients (93%), but myoclonias were only completely controlled in half of the patients at the end of the follow-up period.

Factors identified as being correlated with a lesser efficacy of LTG as add-on medication to VPA in an "analysis of failures" were, the presence of:

1) absences with a mild atonic component in CAE;

2) absences with eyelid myoclonia in CAE, JAE and JME;

3) a longstanding (decades) history of GTCS ;

4) poor compliance.

The lesser response to LTG in some patients with absences with eyelid myoclonia could be another argument that this clinical entity should be considered a separate epileptic syndrome, as proposed by Appleton et al. [23] and Panayiotopoulos [24]. Patients with this syndrome are known to be difficult to treat successfully with classical AED [25]. Nevertheless, in our study LTG was also very efficacious in these patients, which is in concordance with other reports of LTG efficacy in patients with absences with eyelid myoclonia not controlled by VPA and ESM [26-28].

CAE with absences with a mild atonic component is considered a minor variant of typical CAE. The two patients suffering from this condition in our study were the only patients, if patients with compliance problems are disregarded, in whom seizure frequency was not reduced by LTG. This might be interpreted as pharmacological evidence for a different mechanism involved in the generation of absences with an atonic component.

Patients with GTCS on awakening had the lowest response rate, with 3 out of 5 seizure-free patients at the end of the follow-up period. In one patient this was due to poor compliance, in the other patient no specific feature was found apart from a 52-year history of GTCS. Patients with this syndrome are known to be difficult to treat because of their irregular lifestyle and poor compliance [29].

Although EEG photosensitivity was unchanged in 3 photosensitive patients, it increased in 2 and appeared for the first time in 2 of 16 patients on LTG therapy: all photosensitive patients were seizure-free at the end of the follow-up pe-riod, including the three patients with pure photosensitive epilepsy. This long-term observation is especially interesting as LTG has been demonstrated to reduce photosensitivity in 4 and abolish it in 2 out of six patients tested after administration of a single dose of 120 or 240 mg of LTG [30].

Of all the patients in our study taken together, 70% were completely seizure-free at the end of a mean follow-up of 25.5 months. An additional 24% of patients responded partially to LTG, with a greater than 50% reduction in seizure frequency compared to the period prior to LTG introduction. Only 6% of patients did not improve on LTG therapy and in no patient were seizures aggravated.

Myoclonias were the seizure type with the lowest response to LTG. A complete response was obtained in 7 of 14 patients with juvenile myoclonic epilepsy, a complete but transient control in 3 patients with relapse after 6 to 13 months. Myoclonias persisted in 4 patients. Nevertheless, all patients with a relapse of, or persistent myoclonias had a reduction in myoclonia frequency of more than 50%. Similar results were reported by Buchanan [20] in a study on the efficacy of LTG as add-on or monotherapy in 12 patients with JME. In this study, GTCS were controlled in all patients on LTG, but myoclonias on awakening persisted in 3 patients. In 3 patients that were controlled by a combination therapy of VPA and LTG, myoclonias reappeared after an attempt to reduce VPA to more than 50% of its initial dose. The duration of follow-up was not indicated. The long follow-up in our study could be the reason for our finding of a re-emergence of myoclonias after 6 to 13 months, which was resistant to an increase of concomitant VPA.

Weight gain, that had occurred on VPA therapy, was reported to be not reduced after LTG introduction [20], which is contrary to our observation where weight losses of up to 16 kg were documented in 7 out of 18 patients, who had gained weight while on VPA therapy. In the patient charts, these weight losses clearly paralleled the reduction of VPA dosage. As most of these patients were female adolescents, this effect was highly appreciated by the patients and clearly led to a better quality of life. LTG was well tolerated in all but one patient, who developed a rash after LTG was erroneously started at 100 mg/d instead of 12.5 mg/d.

Seizure control was paralleled by a disappearance or major reduction of EEG abnormalities in responding patients. Photosensitivity seemed less influenced by LTG than awake EEG abnormalities at rest or provoked by hyperventilation. A reduction of interictal epileptiform discharges after four months of LTG treatment in a mixed patient group suffering from partial, idiopathic and cryptogenic epilepsies has already been reported [31].

Only few reports on the efficacy of LTG monotherapy in specific IGE syndromes are available [19, 20]. Although there seems to be a synergistic mechanism in the action of VPA and LTG [32, 33], and a combination therapy of both drugs is efficacious and well tolerated in most patients with IGE, it might be preferable to treat patients who have major side effects due to VPA, with LTG monotherapy. In our study, 11 patients were seizure-free on LTG monotherapy with a mean follow-up of 26 months. In 5 patients with JAE and JME, myoclonias or absences reappeared after VPA had been stopped and VPA had to be reintroduced, resulting again in complete seizure control.

Finally, we want to emphasize the importance of this type of open-label, long-term, prospective antiepileptic drug trial in well-defined patient populations with well-characterized seizure types, epilepsy syndromes and etiologies, on the basis of which it is possible to draw conclusions, 1) for the use in clinical practice of a new AED, and 2) on differential pharmacological response profiles to new AED, which will not only influence prescription patterns but which can also be used to generate hypotheses on underlying pathophysiological mechanisms of seizure generation and 3) on seizure and syndrome classification. These conclusions cannot be drawn from the currently performed short-term, double-blind, placebo-controlled trials, which merely have the target of fulfilling licensing criteria, and which are still considered as the "state of the art" trials [34], although they ultimately fail to determine the best use of AED in treating patients [1].

Editorial comment

The point made by the authors of this article that double-blind, short-term controlled studies are more meant to obtain licensing of new drugs than to determine their clinical usefulness in everyday practice is well taken. Comparative, placebo-controlled studies are useful to decide whether a new drug has a real antiepileptic activity, which is an essential basis for later developments. However, much criticism has been directed to the artificial conditions in such trials, to the choice of an end-point (a decrease of 50% in the frequency of seizures often selected may not be of real importance for patients), and to the short duration of many trials. There is clearly a place for long-term, prospective studies, carefully monitored clinically and by EEG and video-EEG recordings, in conditions closer to those of clinical practice. Unfortunately, such studies are subject to multiple biases often difficult to detect and their results have usually been too optimistic.

The development of better new methods of assessment for antiepileptic agents is badly needed. Double-blind controlled studies will continue to play an essential role because of their objectivity but other techniques aiming to reproduce the real conditions of drug use and thus capable of providing answers adapted to patients' needs are clearly required.

Epileptic Disorders will consider the occasional publication of trials designed along these lines provided that they are sufficiently rigorous in their methods and of general interest, despite the criticisms that can legitimately be directed at uncontrolled and/or unblinded studies.

CONCLUSION

Acknowledgement: This study was supported by a grant from Glaxo Wellcome.