Presently, most strategies for development of antiepileptic drugs (AEDs) center around seizure models that are known to respond to presently marketed AEDs. These strategies do not take into account that epilepsy can be a progressive disease. Moreover, region-specific aspects of epileptogenesis are rarely considered when new AEDs are developed. Seizures in the temporal lobe are often difficult to treat. Animal studies on various seizure models in the hippocampus and the entorhinal cortex (EC) suggest that these structures do not a priori produce seizures that are difficult to treat. However, seizure-like events in the EC tend to progress to a state of status epilepticus-like activity that cannot be suppressed by presently marketed AEDs. Loss of gamma-aminobutyric acid (GABA)ergic neurotransmission and increased excitatory synaptic coupling seem to cooperate for induction of this state. Epilepsy induced alterations in the interaction between the EC and the hippocampus may lead to alterations that facilitate precipitation of seizures. Because of the recurrent interaction between the hippocampus and the EC, these seizures may reach an intensity that is no longer controllable by presently available AEDs. Ontogenetic alterations of the circuitry between the EC and the hippocampus, seizure-induced stabilization of synaptic connections overexpressed during ontogenesis, seizure-induced lesions and subsequent rearrangements of internal cell properties, and synaptic arrangements and kindling-like alterations of nerve cell and glial behavior may all be involved in the generation of a neuronal aggregate whose balance between inhibitory and excitatory processes becomes readily disturbed. Strategies for the development of AEDs treating such seizures should suppress hyperactivity and prevent progression of epileptogenesis. AEDs directed against seizures may be effective if they can be given in sufficient concentrations to suppress very intense local seizures.