Antibodies against GluR3 peptides are not specific for Rasmussen's encephalitis but are also present in epilepsy patients with severe, early onset disease and intractable seizures

Abstract

Rasmussen's encephalitis (RE) is a rare condition characterized by drug-resistant seizures, recurrent status epilepticus and progressive lateralized neurological deterioration. There is evidence of autoimmune involvement in the pathogenesis. We investigated the presence of anti-GluR3 antibodies against peptides A and B in patients with RE (n=11), partial and generalized epilepsy (n=85) and other neurological diseases (n=30). The antibodies were specific for epilepsy and are thus not a marker of RE, while particularly high antibody titers characterized a subgroup of non-RE patients with “catastrophic” epilepsy. Antibodies against GluR3B peptide were significantly associated with frequent seizures compared to occasional or drug-controlled seizures.

Introduction

Rasmussen's encephalitis (RE) is a rare progressive form of epilepsy that affects otherwise normal children in the first decade of life Rasmussen et al., 1958, Dulac, 1996. The classic clinical picture is one of severe focal epilepsy with progressive dysfunction of one cerebral hemisphere accompanied by focal neurological signs and cognitive decline. The seizures are refractory to all antiepileptic drugs (Rasmussen and Andermann, 1992). Evidence is accumulating which suggests that RE may be a humoral-mediated autoimmune disease. Thus, epileptic seizures and an inflammatory histopathology of the brain were observed in two rabbits immunized with a portion of the extracellular domain of glutamate receptor 3 (GluR3) fusion protein (Twyman et al., 1995); furthermore, plasma exchange has produced transient improvements in seizure frequency and neurological function in some RE patients Rogers et al., 1994, Andrews et al., 1996, Antozzi et al., 1998. Subsequent studies demonstrated that anti-GluR3 antibodies obtained from GluR3-immunized rabbits promote the in vitro death of cortical neurons by a complement-mediated mechanism (He et al., 1998), and that anti-IgG and anti-membrane attack complex (MAC) antibodies localize on the neuronal cell bodies of sections of cortex resected from RE patients He et al., 1998, Whitney et al., 1999.

However, this emerging picture is complicated by the findings that not all rabbits immunized with the GluR3 fusion protein got sick (Twyman et al., 1995), and that MAC immunoreactivity was present only in a subset of RE patients – those characterized by an active progressive form of RE (Whitney et al., 1999). Furthermore, localized inflammatory reactions have been detected in surgical specimens from some patients who had received temporal lobectomy for refractory epilepsy (McNamara, 1999) raising the prospect of autoimmune involvement in other forms of epilepsy. With regard to RE, there is the challenging conceptual problem as to how autoantibodies might be responsible for progressive destruction of a single cerebral hemisphere.

We carried out a wide-ranging study on the presence of anti-GluR3 antibodies in RE, other forms of epilepsy and other neurological diseases (including autoimmune neurological disorders) to evaluate the specificity and sensitivity of anti-GluR3 antibody testing as a marker for epileptic conditions.