Safety and Tolerability of Perampanel in Cervical Dystonia

Idiopathic cervical dystonia (CD) is the most common form of focal dystonia with a prevalence of approximately 60 cases per million population.(Nutt et al.,1988). Current oral medical treatments for CD have variable efficacy and often with marked side effects. Botulinum toxin injections may be more effective than pharmacological therapies, and are currently the best available therapeutic option. However, repeat injections, administered by a physician trained in this area are required every 3-4 months.(Brans et al.,1996) This can often be difficult and costly for patients. Furthermore, there are subgroups of patients who simply do not respond to this treatment and between 5-20% of patients may become secondary non responders due to the development of blocking antibodies to the botulinum toxin.(Mejia et al., 2005) Thus, new therapeutic options are required.

The neural mechanisms underlying idiopathic dystonia are not well known. Classical basal ganglia circuitry models predict underactivity of the output regions of the basal ganglia, the medial globus pallidus and substantia nigra pars reticulata (;Mitchell et al 1990). In subjects with dystonia undergoing DBS, intraoperative recordings have demonstrated underactivity of the medial globus pallidus (Vitek et al, 1999, Lozano et al 1997). One mechanism responsible for these basal ganglia output changes may be overactivity of corticostriatal glutamatergic pathways, as similar neural mechanism are thought to underlie other hyperkinetic movements (Brotchie 2005). The best studied hyperkinetic movement disorder is levodopa-induced dyskinesia in Parkinson's disease in which dystonia, often of the head and neck, may occur. In animal models of levodopa-induced dyskinesia, increased striatal glutamatergic signaling via alpha-amino-2,3-dihydro-5-methyl-3-oxo-4-isoxazolepropanoic acid (AMPA) receptors has been demonstrated (Perier et al 2002, Silverdale et al 2010). To date there are few validated animal models of idiopathic dystonia. However, one model that has been use for pharmacological studies, and the results extrapolated to idiopathic dystonia, is the paroxysmal dt(sz) dystonic hamster (Loscher and Richter 1998). In this model, intrastriatal and systemic injection of NBQX a selective AMPA receptor antagonist reduced dystonic severity (Richter et al 1993, Sander and Richter 2002, Kohling et al 2004). Other studies have suggested that cerebellar outflow pathways, using AMPA receptors may also mediate dystonic symptoms. Thus the excitatory amino-acid kainite injected into rodent cerebellar vermis resulted in dystonic symptoms, an effect revered by NBQX, suggesting an action on AMPA receptors (Pizoli et al 2002). Thus AMPA receptor antagonists may alleviate dystonia.

To date, clinical studies using glutamate antagonists in CD have been limited due to lack of available drugs. A single 6-week open-label pilot study of the non-selective glutamate antagonist riluzole (50 mg twice a day) in six patients with cervical dystonia (CD) reported a 26% improvement in CD with no side-effects (Muller et al 2002).

The aim of this study is to conduct a multicentre phase I/IIa open label study to determine the safety and tolerability of the AMPA antagonist, perampanel in subjects with primary cervical dystonia. Exploratory analysis will determine effects on dystonia disability and subjective measures including quality of life and global impression of change. The importance of such an initial safety study is due to the lack of knowledge related to the use of this class of drug (AMPA antagonist) in this population of patients. The longer term aim is thus to generate preliminary data for further randomised controlled efficacy studies.