Value of autoantibodies for prediction of treatment response in patients with autoimmune epilepsy: review of the literature and suggestions for clinical management.

The detection of antineural autoantibodies in patients with epilepsy has led to the new concept of "autoimmune epilepsy." A particularly important implication is that knowledge of the antigenic target of the underlying antibody permits prognostic estimates. Patients with antibodies to the potassium channel complex (mostly to its leucine-rich glioma inactivated 1 [LGI1] component) have a high chance of becoming seizure free within days to months upon immunotherapy but less so with antiepileptic drug (AED) treatment alone. Seizures in the setting of antibodies to the N-methyl-d-aspartate receptor also have a high likelihood to remit, again especially with rapid institution of immunotherapy. In contrast to these antibodies to neuronal surface molecules, antibodies directed to intracellular antigens (onconeural antibodies, antibodies to glutamic acid decarboxylase) portend a low likelihood of seizure remission, regardless of the treatments chosen. These outcome differences are probably related to the underlying pathophysiology--with largely reversible functional effects of antibodies to surface antigens and irreversible destructive sequelae (probably caused by T cells) in patients with antibodies to intracellular antigens. With ongoing experience with these conditions, clinical and paraclinical clues to the diagnosis of autoimmune epilepsies are emerging.