Familial Focal Epilepsy with Variable Foci via the DEPDC5 Gene

Pathogenic variants in DEPDC5 cause several epilepsy syndromes, which are characterized by focal seizures arising from discrete areas in the brain. These include: familial focal epilepsy with variable foci (FFEVF), autosomal dominant nocturnal frontal lobe epilepsy (ADNFLE), familial mesial temporal lobe epilepsies (FMTLE), autosomal dominant epilepsy with auditory features (ADEAF), and infantile spasms. FMTLE and ADEAF have been described in only a few nuclear families. It is likely that these two phenotypes are part of the FEVF spectrum and may not be distinct syndromes (Baulac et al. 1993. PubMed ID: 27683934).

FFEVF is a phenotypically heterogeneous form of epilepsy characterized by the major feature of focal seizures. Seizures can originate from different cortical regions of the brain in members of the same family. Common EEG findings include centro-temporal spikes and temporo-posterior sharp waves. Age of onset ranges from infancy to adulthood (Lal et al. 2014. PubMed ID: 24591017; Ishida et al. 2013. PubMed ID: 23542701). In a subset of patients with FFEVF, MRI reveals focal cortical dysplasia with thickening of the cortex and blurring of the grey-white junction (Scheffer et al. 2014. PubMed ID: 24585383). MRI findings within a family can be variable, with some members showing no signs of lesional epilepsy. Intellect is generally preserved, although FFEVF patients have been reported with intellectual disability, autism spectrum, or other psychiatric disorders (Dibbens et al. 2013. PubMed ID: 23542697; Scheffer et al. 2014. PubMed ID: 24585383).

ADNFLE is a partial seizure disorder characterized by seizures that occur during non-REM sleep. Seizure onset is usually between 8 and 12 years, and patients experience predominantly nocturnal seizures with hyperkinetic, tonic, or dystonic features.

A diagnosis of DEPDC5-related epilepsy can provide important prognostic information related to therapy and reproductive planning. Drug-resistant epilepsy is documented in 54-78% of patients (Picard et al. 2014. PubMed ID: 24814846; Baldassari et al. 2019. PubMed ID: 30093711). Favorable surgery outcome has been reported in 60% of individuals in one study (Baldassari et al. 2019. PubMed ID: 30093711). Finally, precision therapy targeting the mTOR pathway has been proposed; however, to date no such medications are available (Myers and Scheffer. 2017. PubMed ID: 28406046)

DEPDC5-related epilepsy—including FFEVF and ADNFLE phenotypes—is inherited in an autosomal dominant manner. Additional genes associated with related epilepsy phenotypes include NPRL2, NPRL3, SCN3A, CHRNA4, CHRNB2, CHRNA2, and KCNT1. The fraction of focal epilepsy attributed to variants in DEPDC5 depends on the epilepsy phenotype studied and ranges from 2-37% of cases (Baulac. 2016. PubMed ID: 27323939). All pathogenic variants in DEPDC5 are believed to be inactivating and include nonsense, frameshift, and missense variants as well as multi-exon copy number losses (Baldassari et al. 2019. PubMed ID: 30093711). The variants are dispersed across the length of the protein, and no clear hotspots have been identified. Based on gnomAD data, DEPDC5 is generally loss-of-function intolerant; however, this is complicated by low quality sequence data that reduces the overall pLI score (Karczewski et al. 2020. PubMed ID: 32461654).

The majority of pathogenic variants are inherited, frequently from unaffected parents. Consistent with that, the penetrance of DEPDC5-related epilepsy is incomplete (approximately 66%, Dibbens et al. 2013. PubMed ID: 23542697; Ishida et al. 2013. PubMed ID: 23542701). However, de novo mutations also contribute to disease (Dibbens et al. 2013. PubMed ID: 23542697; Carvill et al. 2015. PubMed ID: 27066554), and in one case, somatic mosaicism was documented (Baulac et al. 2015. PubMed ID: 25623524). Despite incomplete penetrance, pathogenic variants are very rare in the gnomAD database (allele count <6 and frequency << 0.03%), and some experts recommend leveraging these data to classify more variants as benign (Baldassari et al. 2019. PubMed ID: 30093711).

DEPDC5 (Dishevelled, Egl-10, and Pleckstrin domain-containing protein 5) encodes a GTPase-activating protein (GAP) that binds to NPRL2 and NPRL3 forming the GATOR1 complex. GATOR1 negatively regulates the mTOR phosphorylation pathway (Bar-Peled et al. 2013. PubMed ID: 23723238). Left unchecked, mTOR stimulates cell growth, proliferation, and apoptosis- by acting on the cell cycle, transcription, translation and autophagy (Lasarge and Danzer. 2014. PubMed ID: 24672426). DEPDC5 is not strictly required for cellular viability. However, homozygous knockout mice exhibit preweaning lethality, and conditional knockout of DEPDC5 in mouse neurons causes several phenotypes including seizures (Iffland et al. 2019. PubMed ID: 31625153).

The clinical sensitivity of sequencing the DEPDC5 gene will depend on the patient’s phenotype. Pathogenic DEPDC5 variants have been reported in 12-37% of families with autosomal dominant focal epilepsy (Dibbens et al. 2013. PubMed ID: 23542697; Ishida et al. 2013. PubMed ID: 23542701). DEPDC5 variants were found in 13% (4/30) of cases of ADNFLE in which no pathogenic variants in CHRNA4, CHRNA2, or CHRNB2 were found (Picard et al. 2014. PubMed ID: 24814846).

This test is performed using Next-Generation sequencing with additional Sanger sequencing as necessary.

This test provides full coverage of all coding exons of the DEPDC5 gene, plus 10 bases of flanking noncoding DNA along with other non-coding regions in which pathogenic variants have been identified at PreventionGenetics or reported elsewhere. We define coverage as ≥20X NGS reads or Sanger sequencing.