Clinical manifestations

Presentation and course

Affected family members may present with different types of focal epilepsy, which are, however, invariable within each subject. Frontal lobe seizures are the most frequent manifestation, but they have a different pattern in familial focal epilepsy with variable foci as compared to that observed in autosomal dominant nocturnal frontal lobe epilepsy: the seizures are less frequent at times clustering, focal aware seizures are rare, and daytime seizures as well as focal to bilateral tonic-clonic seizures are more frequently reported (14). Age at seizure onset is variable, usually occurring in the first 3 decades, with 2 peaks at approximately 5 and 25 years of age (75; 74). Although temporal lobe seizures are also commonly found, centroparietal and occipital seizures are less frequent. There is also marked intrafamilial heterogeneity in seizure severity and outcome.

Baldassari and colleagues detailed a large cohort of 73 families in which the probands were diagnosed with GATOR1 complex mutations (07). The age at epilepsy onset ranged from first days of life to 16 years (mean: 4.4 years), with 30% occurring within the first year. Most (68%) patients presented with focal seizures, and almost half of them had predominantly seizures during sleep: 36% with sleep-related hypermotor seizures, 16% with frontal lobe seizures, 32% with focal seizures with undetermined origin, and 10% with infantile spasms.

Drug-resistant seizures developed in 54% of patients, being even more frequent (65%) among patients with sleep-related hypermotor seizures. Interestingly, 60% of these patients had cognitive impairment and/or psychiatric comorbidities, and this was even more frequent (76%) on patients with early onset (≤1 year). In contrast, cognitive dysfunction was reported in 44% (mostly mild to moderate: 20%), whereas learning or language disabilities/delay were reported in 15% probands, also more prevalent among patients with early epilepsy onset.

Prognosis and complications

Familial focal epilepsy with variable foci has an overall benign clinical course, although some patients with refractory seizures have been reported (07). Only 1 individual is known to have undergone a temporal lobectomy, but this was not effective in controlling his seizures (74; 14). No neurologic disabilities are known in this syndrome.

NPRL3 gene mutations can present with drug-resistant seizures, in particular when associated with focal cortical dysplasia (65). In one case report of a proband with familial focal epilepsy with variable foci and stereotactic EEG confirmation of 2 independent seizure foci (left hippocampus and left orbitofrontal cortex), laser interstitial thermal therapy resulted in seizure freedom (02).

Patients with refractory seizures and focal cortical dysplasia type IIa, associated with a DEPDC5 mutation, have been operated with good postoperative results to date (11; 62).

Sudden unexpected death in epilepsy (SUDEP) has been reported. In a large cohort of 73 families, it occurred in 9 individuals: 1 proband had definite sudden unexpected death in epilepsy, and 8 affected relatives had probable sudden unexpected death in epilepsy (mean age at death was 36.8 years, range: 19 to 59 years) (07). Sudden unexpected death in epilepsy was mostly associated with DEPDC5 pathogenic null variants (including 2 sudden unexpected death in epilepsy occurring in 1 family), with only 1 individual carrying a pathogenic loss-of-function variant in NPRL3 gene (07).

In one study, including patients with pathogenic variants in DEPDC5, NPRL2, or NPRL3 and mice models, Bacq and colleagues found no evidence of structural nor functional cardiac damage as contributory factor to sudden unexpected death in epilepsy (04). Studied patients included 3 individuals with DEPDC5 mutations who died from sudden unexpected death in epilepsy and 6 individuals with a family history of sudden unexpected death in epilepsy. Clinical workup including holter, echocardiogram, and electrocardiogram indicated normal cardiac function in these patients. Furthermore, there was no evidence of cardiac injury at autopsy in one postmortem DEPDC5 sudden unexpected death in epilepsy case. The authors used 2 Depdc5 mouse strains: a human HA-tagged Depdc5 strain and a Depdc5 heterozygous knockout with a neuron-specific deletion of the second allele (Depdc5c/-). Simultaneous EEG-ECG recording on Depdc5c/- mice failed to demonstrate any cardiac arrhythmia preceding sudden unexpected death in epilepsy following spontaneous seizures.

Clinical vignette

Xiong and colleagues described the detailed clinical and genetic investigation of 2 large kindreds found to have common ancestors of French-Canadian origin, including 63 family members with seizures or suspected seizures (75; 74). A description of these individuals is out of the scope of this clinical summary and can be found in these previous publications. For illustrative purposes, we provide the details of the proband of 1 of these families.

At the time of identification, the proband of pedigree number 22 was a 13-year-old boy with refractory seizures since the age of 5. Previously used antiepileptic drugs included carbamazepine, phenytoin, clobazam, lamotrigine, valproate, primidone, and vigabatrin. Most often, seizures occurred in the morning or late in the evening, without clustering, once every 1 or 2 weeks. He had 2 types of seizures: (1) deep breathing and lip smacking followed by head turn to the right, tonic flexion of the left arm and right hand automatisms (tonic extension of the left leg can occur), and mild postictal confusion; (2) occurring during sleep with awakening, trunk movements (rocking back and forth in bed) with occasional pelvic thrusting, scissor-like movements, and groaning sounds with deep breaths.

His general physical and neurologic examinations were normal. Video-EEG monitoring revealed bilateral frontal onset of seizures with right-sided predominance. Interictal EEG showed an epileptiform disturbance over the right hemisphere, involving the fronto-centro-parietal and temporal regions, with spike and sharp wave activity observed mainly in the right frontal and temporal regions. MRI, including volumetric studies, was normal, although the right hippocampus appeared bigger than the left. Dysfunction over the right temporal region was revealed by 1H-MRS (reduction of N-acetylaspartate) and by [18F]FDG-PET (hypometabolism). Neuropsychological evaluation demonstrated no cognitive dysfunction, a full-scale IQ of 118, and suggestion of left-sided speech representation.

Biological basis

Etiology and pathogenesis

The etiology of familial focal epilepsy with variable foci started to be clarified with the description of mutations in the DEPDC5 gene in ch 22q in approximately 37% of families studied (10). However, as such mutations have also been found in smaller families including those with autosomal dominant nocturnal frontal lobe epilepsy, rolandic epilepsy, temporal lobe epilepsy, epileptic spasms, as well as patients with no family history of seizures, their specificity in how they lead to the unique presentation in familial focal epilepsy with variable foci remains to be further elucidated (34; 41; 46; 58; 11; 17; 22; 59; 62; 69; 61; 71; 76).

The identification of mutations in genes encoding other components of GATOR1 complex, which includes not only DEPDC5 but also NPRL2 and NPRL3, suggests that the resulting pathogenesis and pathophysiology might derive from dysregulation of the mTOR pathway, as GATOR1 complex is involved in the inhibition of the mTOR complex 1 (06; 10).

Approximately 183 patients have been reported with a GATOR1 mutation/variant, DEPDC5 being the most frequent (155, 85%), and NPRL2 the less prevalent (10 patients) (07). Baldassari and colleagues highlight that this is likely due to DEPDC5 transcript being longer (5551bp) than NPRL2 (1700bp) and NPRL3 (2881bp), and the latter 2 have been less frequently tested due to their more recent discovery (07). NPRL2 and NPRL3 mutations have also been found in familial and nonfamilial cases, with or without focal cortical dysplasia (39; 61; 65; 72; 07; 45; 70).

The DEPDC5 gene encodes a protein expressed in neurons throughout brain development (46). The protein has a DEP domain (Dishevelled, Egl-10, and Pleckstrin domain) that is found in proteins involved in G-protein signaling pathways.

Most of the 140 different variants identified so far correspond to loss-of-function mutations (67%), 24 of which were found in unrelated cases suggesting mutational hotspots or founder effects. Less frequently, missense (27%), splice-region variants (4%), and in-frame indels (1%) have been reported (07).

A DEPDC5 splice mutation has been identified in a Taiwanese kindred with familial focal epilepsy with variable foci (71), and also in 1% of sporadic and mostly nonlesional focal epilepsy patients (N=293) investigated with targeted resequencing had possibly pathogenic variants in the DEPDC5 gene.

From the French Canadian familial focal epilepsy with variable foci cohort, DEPDC5 mutations were found in 5% of familial and sporadic focal epilepsy cases (4/79), with identification of a founder mutation specific to this population (46). DEPDC5 loss-of-function mutations were found in 13% of these families with a presentation of autosomal dominant nocturnal frontal lobe epilepsy.

Although there was suggestive linkage to chromosome 2q in the original family described by Scheffer and colleagues, this was later not confirmed, and all other families tested showed definite linkage to chromosome 22q (64). In the series of 19 autosomal dominant families with focal epilepsy described by Picard and colleagues, 4 had familial focal epilepsy with variable foci (57). These families have been reevaluated and showed linkage to chromosome 22q12, with mutations in the DEPDC5 gene found in 37% of them (34).

The identification of DEPDC5 mutations in patients with malformations of cortical development is a major discovery, as clinical observation of focal cortical dysplasia in the context of familial epilepsies had been noted (11; 22; 62). In addition to familial focal cortical dysplasia associated with DEPDC5 mutations, DEPDC5 mutations have also been found in patients with sporadic focal cortical dysplasia hemimegalencephaly and pachygyria (22; 18).

In a large phenotyping study including 303 families with epilepsy, 7 out of 62 (11.3%) kindreds with focal epilepsy were classified as familial focal epilepsy with variable foci, many of which had been tested negative for DEPDC5 mutations (26). DEPDC5 was the only gene with study-wide significance in an exome sequence study derived from the Epilepsy Phenome/Genome Project and Epi4K Consortium comparing individuals with familial genetic generalized epilepsy (N=640), familial nonacquired focal epilepsy (N=525), and 3877 controls (26). In addition to DEPDC5, ultra-rare deleterious variants were identified in 4 other genes previously known to be associated with epilepsy (LGI1, PCDH19, SCN1A, and GRIN2A), altogether contributing to the risk of epilepsy in approximately 8% of individuals with familial nonacquired focal epilepsy.

The mechanisms by which mutations in the DEPDC5 gene can present as different types of focal epilepsy within the same family are intriguing (09). The same questions remain for the presence and type of underlying malformation of cortical development. The two-hit somatic mutation hypothesis has been demonstrated in 2 independent studies, which included in utero electroporation to create focal somatic DEPDC5 deletion in the rat embryonic brain (32; 60). Interestingly, this resulted in spontaneous seizures with focal pathological and electroclinical features as seen in patients with focal cortical dysplasia type IIa (32). This second hit mutation might be cell-specific, such as in dysmorphic neurons of focal cortical dysplasia type IIa (42).

Many studies on surgical brain tissue have provided important information to date. Analysis of freshly frozen tissue from surgery in a patient with refractory seizures demonstrated a mutation gradient with a higher rate of mosaicism in the seizure-onset zone than in the surrounding epileptogenic zone (60).

A role for DEPDC5 in development of epileptogenesis by affecting excitatory synapses and increasing expression of glutamate receptors could underlie the mechanism through which these mutations relate to epilepsy phenotypes and mTOR hyperactivation (21).

In a cohort of surgical specimens derived from 80 children with neuropathological diagnosis of focal cortical dysplasia and hemimegalencephaly, targeted gene sequencing identified variant in 29% of focal cortical dysplasia type I patients and 63% of focal cortical dysplasia type II or hemimegalencephaly patients (07b). Germline, somatic, and two-hit loss-of-function variants in DEPDC5 were found exclusively in the latter group of patients.

Pathogenic mutations were identified in 31.9% of 446 tissues samples from 232 epilepsy patients with various underlying neuropathological diagnosis, including DEPDC5 (66). Targeted panel deep sequencing on paired blood and brain-derived genomic DNA from operated patients with bottom of the sulcus dysplasia revealed pathogenic germline DEPDC5 and NPRL3 variants in 2 out of 20 patients (43).

Animal models provide useful insight into pathophysiological mechanisms of these gene alterations. A knockout DEPDC5 mouse generated using targeted CRISPR mutagenesis revealed homozygous phenotypes that support mTORC1 hyperactivation as a pathogenic mechanism underlying DEPDC5 loss of function in humans (33). Heterozygous mice had normal phenotypes. The mTORC1 hyperactivation mechanism has been further demonstrated by Ribierre and colleagues, who performed a single intraperitoneal injection of rapamycin into pregnant dams at E15, preventing the neuronal migration disorder seen in DEPDC5 knockout embryos (ie, cells retained into the ventricular and subventricular zones at E18.5) (60).

Mice in which NPRL2 and NPRL3 were conditionally deleted from the dorsal telencephalon present with spontaneous seizures and dysmorphic enlarged neuronal cells with increased mTORC1 signaling, similar to Depdc5-cKO mice (35). Chronic postnatal administration of rapamycin had a therapeutic effect on clinical presentation (prolonging survival period and inhibiting seizure occurrence), which was, however, short-lasting after cessation of rapamycin as compared to that observed in Depdc5-cKO mice.

The finding of DEPDC5 and NPRL2 mutations in patients who had definite or probable sudden unexpected death in epilepsy represents another important clinical association that warrants further understanding on the pathogenesis of GATOR1 genes (51; 05; 72; 07). In 3 DEPDC5 knockout mice, a single seizure was followed by sudden death, further indicating that focal brain mosaic knockout of a non-channelopathy gene in mice correlates with the few observations of sudden unexpected death in epilepsy in humans (60). Premature death was observed in Depdc5-Emx1-Cre conditional knockout mice, which is a novel model of Depdc5 deficiency with severe epilepsy and macrocephaly, generated by conditional deletion of Depdc5 in dorsal telencephalic neuroprogenitor cells (38). In these animals, mTORC1 inhibition with rapamycin significantly improved survival and seizures (38).

Epidemiology

Familial focal epilepsy with variable foci kindreds have been identified in Australia, Canada, China, Spain, Holland, Morocco, France, Argentina, and Taiwan. Three French-Canadian families originating from the region around Quebec City share the same haplotype on chromosome 22q, and molecular studies showed a recurrent p.R843X protein-truncating mutation segregating in 1 of these families, with suggestion of an ancestral allele (75; 74; 14; 46). Due to the variable seizure pattern among affected family members and the usually benign clinical outcome, familial focal epilepsy with variable foci might often be underdiagnosed, as is the case for other familial focal epilepsy syndromes.

Prevention

Genetic counseling and prenatal diagnosis would now be feasible.

Differential diagnosis

Sporadic benign focal epilepsies constitute the main clinical differential diagnosis, but unless tested, it is impossible to tell whether these individuals may have mutations in the same gene or genes as patients with familial focal epilepsy with variable foci. It should be noted that some sporadic patients could, in fact, be part of a familial focal epilepsy with variable foci kindred in which the other family members are only mildly affected and sometimes unrecognized. In the case of sporadic French-Canadian patients, however, identification of the same haplotype on chromosome 22q as in the previously reported French-Canadian families constitutes a high probability for a positive diagnosis of familial focal epilepsy with variable foci (75; 74; 14). Finding of mutations in GATOR1 complex genes can support the diagnosis of familial focal epilepsy with variable foci, but mutations can also be found in patients without a familial epilepsy syndrome.

Among the genetically determined focal epilepsies, the most important differential diagnosis is autosomal dominant nocturnal frontal lobe epilepsy or sleep related hypermotor seizures because frontal lobe seizures are the most frequent type found in affected family members with familial focal epilepsy with variable foci. For autosomal dominant nocturnal frontal lobe epilepsy, 3 loci and 2 genes have been identified: ENFL1 (chr 20q13.2), with 4 different mutations in the CHRNA4 gene, coding for the alpha 4 subunit of the neuronal nicotinic acetylcholine receptor (AchR); ENFL2 (chr 15q24, gene still unknown); and the CHRNB2 gene on the ENFL3 locus (chr 1q), coding for the beta 2 subunit of the AchR (55; 56; 54; 68; 67; 31; 20; 28; 44). However, most families (88%) with autosomal dominant nocturnal frontal lobe epilepsy do not map to any of these loci and do not have mutations in either the CHRNA4 or CHRNB2 genes. Mutations in the sodium-gated potassium channel gene KCNT1 have also been identified in more severe phenotypes (30).

Aridon and colleagues described a mutation in the CHRNA2 gene associated with nocturnal seizures, fear, and nighttime wandering (03). However, Gu and colleagues did not find mutations in the CHRNA2 gene in 47 families with autosomal dominant nocturnal frontal lobe epilepsy (29). The report of autosomal dominant nocturnal frontal lobe epilepsy kindreds with DEPDC5 or NPRL3 mutations adds to the complexity of the differential diagnosis (58; 39).

In the absence of genetic conformation, the differential diagnosis between familial focal epilepsy with variable foci and autosomal dominant nocturnal frontal lobe epilepsy should be based on the clinical and EEG characteristics of the frontal lobe seizures as well as family history of seizure patterns that indicate foci outside the frontal regions in familial focal epilepsy with variable foci.

The familial temporal lobe epilepsies, both familial mesial temporal lobe epilepsy and familial lateral temporal lobe epilepsy should also be considered in the differential diagnosis if 1 or more affected individuals in the family presents clinical and EEG features of temporal lobe epilepsy. A confirmed extratemporal focus on a family member would exclude the diagnosis of familial mesial temporal lobe epilepsy or familial lateral temporal lobe epilepsy and could suggest familial focal epilepsy with variable foci. Genetic investigation of mutations in the LGI1 gene on chromosome 10q can suggest a possible diagnosis of familial lateral temporal lobe epilepsy (36; 50). LGI1 mutations have been found in about half of the families investigated (13). Mutations in RELN gene on chromosome 7q have been found in 17% of kindreds that tested negative for LGI1 mutations (19; 47). However, the observation of DEPDC5, NPRL2, and NPRL3 mutations in temporal lobe epilepsy probands of mesial and lateral forms further complicates the differential diagnosis based on genetics (59; 69; 15; 61).

An updated review of these familial epilepsy syndromes can be found in the dedicated MedLink clinical summaries for Genetic epilepsy with febrile seizures plus, Familial lateral temporal lobe epilepsy, Familial mesial temporal lobe epilepsy, and Autosomal dominant nocturnal frontal lobe epilepsy.

Diagnostic workup

A detailed clinical evaluation with clear description of the seizures by the patient and close relatives is the first important requirement for diagnosis of familial focal epilepsy with variable foci, as for any other epilepsy patient.

EEG recordings and, whenever possible, prolonged video-EEG monitoring can help in the definition of the epileptic focus. Interictal epileptiform abnormalities have been recorded in the EEG of 83% of probands, reaching 90% among patients with hypermotor seizures (07).

Structural and/or functional neuroimaging abnormalities, most often revealing an underlying malformation of cortical development, have been found in 38% of probands. These include focal or hemispheric malformations, with focal cortical dysplasia in approximately 20% of probands, and rarely, hippocampal atrophy. The presence of an MRI lesion suggestive of a malformation of cortical development might lead to the definition of familial epilepsy with focal cortical dysplasia. In contrast, kindreds with solely MRI negative and seizure-free patients have also been described (01).

Genetic testing can now be performed and might show DEPDC5, NPRL2, or NPRL3 mutations. The clinical yield of diagnostic exome sequencing for genetic screening of main genes involved in focal epilepsies is promising (40). In a cohort of 112 patients, 12% showed diagnostic variants, most of which (69%) were GATOR1 genes.

In a study evaluating the yield of whole exome sequencing screening of targeted gene analysis in clinical practice, 40 consecutive patients who had MRI-negative focal epilepsy and a family history of seizures in at least 1 first- or second-degree relative have been evaluated for 64 epilepsy genes (53). Five (12.5%) patients had a pathogenic (or likely pathogenic) variant in SCN1A, DEPDC5, PCDH19, GABRG2, or NPRL2 genes, and at least in 1 patient these results determined an important change in the treatment decision, leading to significant improvement of the epilepsy.

Management

Treatment should be based on the patient’s response to antiseizure drugs, and the rationale is similar to that in nonfamilial patients with focal epilepsies. Usually, patients with familial focal epilepsy with variable foci are well controlled with small doses of the same medications indicated in other focal epilepsies, and they may remit spontaneously.

Rather, some patients might be found to have familial epilepsy with focal cortical dysplasia, and surgery should be equally considered for those that are refractory to medical treatment (11).

In a study evaluating the role of rare genetic variants in pharmacoresistance, no gene was found to reach genome-wide significance, with DEPDC5 showing, however, a potential association with resistance to antiseizure medications (73).