Page 47 - ILAE_Lectures_2015

Progress has been made also in the focal epilepsies. Perhaps most interesting, and what may
possibly emerge as the most common genetic cause in familial focal epilepsies, is the
discovery of mutations in DEPDC5 in several familial and sporadic epilepsy phenotypes,
including familial focal epilepsy with variable foci, familial temporal lobe and autosomal
dominant nocturnal frontal lobe epilepsies, and rolandic epilepsy54-56. These findings are of
especial interest as DEPDC5 is part of the mTOR pathway57, activity within which can be
manipulated using the existing drug rapamycin. DEPDC5 mutations have also been shown
in epilepsies with developmental malformation58. Somatic mutations in MTOR itself have
been reported in focal cortical dysplasia and hemimegalencephaly59. Mutations in the ion
channel gene KCNT1 have been reported in malignant migrating partial seizures of infancy
and severe autosomal dominant nocturnal frontal lobe epilepsy60,61. Most focal epilepsies,
however, remain genetically unexplained.

Finally, there are of course also the epilepsies across the spectrum with well-established
genetic causation. These epilepsies include those associated with developmental structural
abnormalities, neurocutaneous disorders, chromosomal disorders established well before the
aCGH era, several PMEs, neurometabolic disorders, mitochondrial cytopathies, the focal
epilepsies, autosomal dominant frontal lobe epilepsy and lateral temporal lobe epilepsy,
Dravet syndrome, Rett and related syndromes. Several excellent reviews have been published
on these conditions. It is also worth noting that not all cases with phenotypes similar or related
to these epilepsies have actually been solved and efforts continue to explain these. Just to
give two examples, new genes have been identified for Dravet syndrome (such as CHD262)
and for polymicrogyria (such as CCND263) in various settings.

Beyond the discovery of genetic causes of specific types of epilepsy, other aspects of the
epilepsies are also being investigated. Given the breadth of phenotypic variation seen in some
otherwise characteristic epilepsies, there has been much interest in genetic modifiers of
phenotype. Identifying modifiers is challenging as many factors other than genetic variation
may play a role. Animal models have been explored from this perspective, with evidence for
example that mutations in different genes may influence the epileptic phenotype64. In
humans, SCN9A has been proposed as a modifier of the Dravet and GEFS+ phenotypes65.
Some have considered microdeletions to be modifiers within genetic generalised epilepsy
phenotypes66. Taking this concept further, network disruption has been proposed in mesial
temporal lobe epilepsies with hippocampal sclerosis67. Cause and effect can be difficult to
disentangle in such studies, and the standards for proof are yet to be clarified in this area. The
provocative effect of intermittent photic stimulation in precipitating seizures has been a topic
of genetic research for many years. The area is complex, with definitions and protocols
varying between studies, sites and publications. Taking broad common phenotypes into
account, and based on the observation that photosensitivity is frequently present in epileptic
encephalopathy due to CHD2 mutation or deletion68, it was shown that CHD2 mutation was
also present in a small proportion of people with photosensitivity and more common
epilepsies, and was present in 3/36 patients with the syndrome of eyelid myoclonia with
absences69.

Treatment genomics in the epilepsies remain a challenge. In keeping with most trials, most
studies have been drug-centred. Variants significantly increasing the risk of severe or mild
rash on exposure to carbamazepine have been identified in the HLA system, with HLA-
B*1502 being a major risk in populations of South Asian extraction70 and HLA-A*3101 in
people of European extraction71. Screening for the B*1502 variant has been shown to be cost-
effective in a south Asian population72. A systematic review has shown that HLA-B*1502 in
Asian patients is associated with a pooled odds ratio of 113.4 for severe carbamazepine-
induced reactions (Stevens-Johnson syndrome and toxic epidermal necrolysis), and that 461
patients would need to be screened for HLA-B*1502 to prevent one episode of such a severe
reaction73. For HLA-A*3101, which is more broadly associated with cutaneous