Clinical Description
Autosomal dominant epilepsy with auditory features (ADEAF) is characterized by focal epilepsy not caused by a previous illness or injury, with auditory symptoms and/or receptive aphasia as prominent ictal manifestations. Age at onset has ranged from four to 50 years in previously reported families [Winawer et al 2000, Brodtkorb et al 2002, Winawer et al 2002, Michelucci et al 2003, Michelucci et al 2013], but is usually in adolescence or early adulthood. The prominent auditory symptoms and aphasia are thought to reflect a localization of the epileptogenic zone in the lateral temporal lobe; accordingly, ADEAF is also known as autosomal dominant lateral temporal epilepsy (ADLTE).
Epilepsy. Affected individuals have focal to bilateral tonic-clonic seizures, usually accompanied by focal aware or focal impaired-awareness seizures, with auditory symptoms as a major focal aware seizure manifestation occurring in around two thirds of affected individuals. Some individuals have seizures precipitated by specific sounds, such as a telephone ringing [Michelucci et al 2003, Michelucci et al 2004, Michelucci et al 2007].
Although most individuals in families with ADEAF have focal epilepsy, idiopathic generalized epilepsy was reported in four individuals with LGI1 pathogenic variants in two previously reported families [Ottman et al 2004]. The occurrence of idiopathic generalized epilepsies in these families may be explained either as an effect of LGI1 on the risk for idiopathic generalized epilepsy, or by the co-occurring pathogenic variant in these families of another (unidentified) gene that specifically influences risk for idiopathic generalized epilepsy.
Febrile seizures do not occur with increased frequency in ADEAF.
Auditory symptoms. The most common auditory symptoms are simple unformed sounds such as humming, buzzing, or ringing. Less frequently, other types of auditory symptoms occur, including complex sounds (e.g., specific songs or voices) or distortions (e.g., volume changes). Negative auditory symptoms, such as sudden decrease or disappearance of the surrounding noises, are reported by a minority of affected individuals.
Aphasia. Another distinctive feature is ictal receptive aphasia (i.e., sudden onset of an inability to understand language, in the absence of general confusion). Ictal aphasia was the most prominent symptom in one large Norwegian family with an LGI1 pathogenic variant [Brodtkorb et al 2002, Brodtkorb et al 2005a] (although auditory symptoms also occurred) and in a small Japanese family [Kanemoto & Kawasaki 2000]. Aphasia has also been reported in other families with LGI1 pathogenic variants [Michelucci et al 2003, Ottman et al 2004, Di Bonaventura et al 2009].
Other ictal symptoms. In families with ADEAF, affected individuals also have other ictal symptoms, either in isolation or accompanying auditory symptoms or aphasia. These occur less frequently than auditory symptoms and include other sensory symptoms (visual, olfactory, vertiginous, or cephalic) as well as motor, psychic, and autonomic symptoms [Poza et al 1999, Winawer et al 2000, Winawer et al 2002, Michelucci et al 2003, Hedera et al 2004, Ottman et al 2004, Michelucci et al 2013, Dazzo et al 2015b].
Non-epileptic manifestations associated with ADEAF on rare occasion include the following:
Behavioral problems (e.g., explosive violent behaviors, impulsiveness) and depression (with suicide attempts) have been reported in single pedigrees [
Chabrol et al 2007,
Kawamata et al 2010]. However, a systematic study investigating a possible shared genetic susceptibility to epilepsy and depression in families with an
LGI1 pathogenic variant did not find such an association; rather, the depression appeared to be related to the epilepsy or anti-seizure treatment [
Heiman et al 2010].
Migraine segregating with occipito-temporal epilepsy resembling ADEAF has been described in one family [
Deprez et al 2007].
Prognosis. The clinical course of ADEAF is usually benign. The following are offered as examples.
In a series of 34 affected individuals in seven Spanish and Italian families, focal to bilateral tonic-clonic seizures occurred only once or twice per year. The frequency of focal aware or focal impaired-awareness seizures ranged from twice per year to several times per month. After initiation of medical therapy, seizures were well controlled by any of a variety of medications (carbamazepine, phenobarbital, or phenytoin), sometimes at low doses [
Michelucci et al 2003].
In a Norwegian family with prominent ictal aphasia, all individuals had been free from focal to bilateral tonic-clonic seizures for two or more years, and focal aware seizures occurred infrequently in most individuals. However, two family members with epilepsy died suddenly in their sleep, both at age 28 years; a relationship to seizures was suspected but could not be confirmed [
Brodtkorb et al 2002].
EEG. Interictal (routine and sleep-deprived) EEGs may be normal in persons with ADEAF; however, epileptiform interictal EEG abnormalities are found in up to two thirds of affected individuals [Poza et al 1999, Winawer et al 2000, Brodtkorb et al 2002, Winawer et al 2002, Fertig et al 2003, Michelucci et al 2003, Pizzuti et al 2003, Hedera et al 2004, Ottman et al 2004, Pisano et al 2005]. Interestingly, a left predominance of the abnormalities has been observed in some clinical series [Michelucci et al 2003, Di Bonaventura et al 2009].
Ictal EEGs have been reported in three persons [Winawer et al 2002, Brodtkorb et al 2005a, Di Bonaventura et al 2009]. One of these showed left mid- and anterior temporal onset [Winawer et al 2002], and another onset in the left frontotemporal region with bilateral and posterior spreading, documented during a video-recorded aphasic seizure [Brodtkorb et al 2005a]. The third was recorded during a prolonged seizure cluster lasting several hours in an individual with prominent ictal aphasia; the EEG pattern consisted of low-voltage fast activity followed by delta activity and rhythmic sharp waves located in the anterior and middle left temporal regions [Di Bonaventura et al 2009].
Findings from magnetoencephalography (MEG) with auditory stimuli showed significantly delayed peak 2 auditory evoked field latency in individuals with LGI1 pathogenic variants [Ottman et al 2008]. Another study using MEG detected significantly large N100m signals in three of five individuals, contralateral to the auditory stimulation [Usui et al 2009].
Neuroimaging. Findings from routine neurologic examination and routine clinical imaging (MRI or CT) are normal.
An interictal single-photon emission computed tomographic scan in one person identified hypoperfusion in the left temporal lobe [Poza et al 1999].
A left lateral temporal lobe malformation was identified through high-resolution MRI in ten individuals in a Brazilian family with an LGI1 pathogenic variant [Kobayashi et al 2003]. However, other studies using high-resolution MRI in families with LGI1 pathogenic variants have not confirmed this finding [Tessa et al 2007, Ottman et al 2008].
Diffusion tensor imaging identified a region of increased fractional anisotropy in the left temporal lobe in individuals with an LGI1 pathogenic variant [Tessa et al 2007].
In functional MRI with an auditory description decision task, persons with epilepsy in families with an LGI1 pathogenic variant had significantly less activation than controls [Ottman et al 2008]. These results suggest that individuals with ADEAF have functional impairment in language processing.
Other investigations. Asymmetry of long-latency auditory evoked potentials (with reduced left N1-P2 amplitudes) was shown in the Norwegian family with aphasic seizures [Brodtkorb et al 2005b]. Abnormal phonologic processing was demonstrated in four persons in a Sardinian family by means of a fused dichotic listening task [Pisano et al 2005]. The above data, though based on a small sample size, would appear to suggest the existence of some structural abnormalities in the lateral temporal neuronal network.
Penetrance
Estimates of penetrance in studies of families with ADEAF range from 54% to 85% [Ottman et al 1995, Poza et al 1999, Ottman et al 2004, Wang et al 2006]. This variability may in part result from the use of different statistical models.
LGI1. Based on analysis of obligate heterozygotes in 24 published families, penetrance of LGI1 pathogenic variants was estimated at 67% (95% CI 55%-77%) [Rosanoff & Ottman 2008].
More recently, in a study of 33 families in which probands were excluded, penetrance for epilepsy was estimated at 61% in ten families with an LGI1 pathogenic variant and 35% in families without an identified pathogenic variant, suggesting that inheritance may be complex in some families [Michelucci et al 2013].
All of these estimates are likely to be inflated by ascertainment bias, since they are based on families selected for study because they comprised many affected individuals.
RELN. Twenty (60%) of 33 individuals heterozygous for a RELN pathogenic variant (from 7 families) had epilepsy [Dazzo et al 2015a].
MICAL1. Penetrance is unknown.