A new generation of
antiepileptic drugs has emerged in the past 10 years, including gabapentin,
lamotrigine, felbamate, and vigabatrin. These are said to be valuable adjuncts to
the first line drugs when epilepsy is inadequately controlled. No agreement
exists on the criteria that define treatment failure. One view is that a
treatment has failed when seizures are unacceptably frequent despite plasma drug
concentrations in the "therapeutic range." However, this range is defined as the
range of concentrations at which most patients have a size able reduction in the
frequency of seizures without substantial dose dependent side effects. Since some
patients require and tolerate larger doses, treatment failure might be better
defined as inadequate efficacy at the highest tolerated dose. Drug treatment
for epilepsy is usually long term and should therefore be as simple as possible.
Ideally, patients should start treatment with a single drug.(1) In the countries
where they are now available, vigabatrin and lamotrigine were approved on the
basis of supplementary treatment. Randomised controlled trials were performed in
patients whose epilepsy was inadequately controlled by standard doses of a first
line drug.(2-4) This dose was continued along with either the new antiepileptic
drug or a placebo. A systematic review of 28 of such randomised placebo
controlled "add on" trials appears in this week's issue (p 1169).(5) It concludes
that the new antiepileptic drugs are significantly better than placebo in
reducing seizure frequency, but that comparative trials will be needed to see if
one drug is better than another. However, trials of supplementary drugs cannot
measure the efficacy of a new drug. What they assess is the overall effect of the
combination, which may be due to simple additive effects but which may also
result from synergistic or even antagonistic effects.
The trials of
vigabatrin and lamotrigine give other grounds for concern. The groups were not
homogeneous at baseline: some patients were receiving a single drug and some
multiple drugs, and some met one and some the other of the two criteria for
treatment failure cited above. The drug combinations tested were also highly
diverse. Most important, the main end point used was a reduction in the frequency
of attacks by 50% or more; no account was taken of the clinical benefit
experienced by the patients. Yet in a recent trial of lamotrigine no correlation
was found between reductions in the frequency and severity of seizures and
patients' wellbeing.(6)
None of the trials lasted more than three years. As
these new antiepileptic drugs are costly, they should not be prescribed routinely
without data on how many patients actually benefit and for how long. Also in this
issue (p 1184) Walker et al show that most patients who initially respond to
supplementary lamotrigine or vigabatrin gradually abandon the new drugs.(7)
Studies will be needed to determine the reasons, which might include reduction in
efficacy, side effects, or cost.
More detailed questions remain unanswered.
The initial assessment file on vigabatrin left some doubt about potential side
effects since studies on animals had reported ocular and neurological
toxicity.(5) In France vigabatrin was initially reserved for use in hospital
neurology units, which had to include the patients in a cohort study to identify
any such toxicity. Unfortunately, no conclusions can be drawn from data published
so far because of methodological problems (Mauguiere et al, personal
communication, 1995).
Drug regulatory agencies should not be approving new
drugs for which no benefit has been proved for patients using relevant outcome
measures. In the case of the new antiepileptic drugs I believe that problems with
the design of the trials have undermined the reliability of the available data.
When marketing approval is authorised on the basis of relatively short trials (so
that patients can benefit rapidly) I believe the assessment should be continued
to determine the cost effectiveness and risk-benefit ratios in the long term. The
regulatory agencies should reassess the files on these new antiepileptic drugs at
regular intervals, and meanwhile their licences should include a requirement for
a programme of continuing evaluation.
GILLES MIGNOT Clinical
pharmacologist
La revue Prescrire (English edition, Prescrire
International)
BP 459,
75527 Paris Cedex 11,
France
References
1 Shorvon S D. Medical assessment and treatment of chronic epilepsy. BMJ
1991;302:363-6.
2 Vigabatrin. Prescrire International
1993;2(5):3-4.
3 Vigabatrin and childhood epilepsy. Prescrire
International 1993;2(5):22-3.
4 Lamotrigine. Prescrire
International (in press).
5 Marson A G, Kadir Z A, Chadwick D W. The new antiepileptic drugs: a systematic review of their
efficacy amnd tolerability. BMJ 1996;313:1169-74.
6
Chadwick D. Measuring antiepileptic therapies: the patient vs the physician view
point. Neurology 1994;44 (suppl 8): S24-5.
7 Walker M C, Li L
M, Sander J W A S. Long term use of the new anti-epileptic drugs, lamotrigine and
vigabatrin in severe refractory epilepsy. BMJ 1996;313;1184-5.
