Download Obtaining genetic testing in pediatric epilepsy

CRITICAL REVIEW AND INVITED COMMENTARY
Obtaining genetic testing in pediatric epilepsy
*†Margie A. Ream, and *†Anup D. Patel
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
SUMMARY
Margie Ream is an
assistant professor of
pediatric neurology at
Nationwide Children’s
Hospital and the Ohio
State University in
Columbus, Ohio.
The steps from patient evaluation to genetic diagnosis remain complicated. We discuss some of the genetic testing methods available along with their general advantages
and disadvantages. We briefly review common pediatric epilepsy syndromes with
strong genetic association and provide a potentially useful algorithm for genetic testing in drug-resistant epilepsy. We performed an extensive literature review of available information as it pertains to genetic testing and genetics in pediatric epilepsy. If a
genetic disorder is suspected as the cause of epilepsy, based on drug resistance, family
history, or clinical phenotype, timely diagnosis may reduce overall cost, limit the diagnostic odyssey that can bring much anxiety to families, improve prognostic accuracy,
and lead to targeted therapy. Interpretation of complicated results should be performed only in collaboration with geneticists and genetic counselors, unless the ordering neurologist has a strong background in and understanding of genetics. Genetic
testing can play an important role in the care provided to patients with epilepsy.
KEY WORDS: Exome sequencing, Next generation sequencing, Diagnosis.
Epilepsy is a common neurologic disorder affecting
1–1.5% of the world’s population and is more commonly
diagnosed in children than adults. Historically, up to 70% of
epilepsy etiology was “idiopathic,”1 but with the advent
of widely available genetic testing, an etiology in subsets of
epilepsy patients (for example, generalized epilepsy with
developmental delay) may be obtainable in >30%.2 Updated
International League Against Epilepsy (ILAE) classification emphasizes genetic classification while downplaying
the utility of the designation “idiopathic.”3 However, the
steps from patient evaluation to genetic diagnosis remain
complicated. Later we discuss some of the genetic testing
methods available along with their general advantages and
disadvantages. Then we briefly review common pediatric
epilepsy syndromes with strong genetic associations and
Accepted July 22, 2015; Early View publication September 8, 2015.
*Nationwide Children’s Hospital, Columbus, Ohio, U.S.A.; and †The
Ohio State University College of Medicine, Columbus, Ohio, U.S.A.
Address correspondence to Anup D. Patel, ED 533, 700 Children’s
Drive, Columbus, OH 43205, U.S.A.
E-mail: [email protected]
Wiley Periodicals, Inc.
© 2015 International League Against Epilepsy
provide a potentially useful algorithm for genetic testing
(Table 1).
Genetic Testing Background and
Methods
The human genome contains approximately 3 billion
DNA bases that encode approximately 25,000 genes spread
across 23 chromosomes. Pathogenic changes can occur at
many levels. Two major classes of variation are seen: copy
number variation (CNV) in which there are duplications or
deletions of portions of DNA, and single nucleotide variant
(SNV) in which a single base position has been altered.
Once a sequence is determined, it is compared to a reference
sequence, which is based on consensus of thousands of individuals of different ethnicities to determine if the sequence
in question is normal or varies from the expected standard
reference. Variants are classified by their pathogenicity,
either based on previous reports or on analysis by in silico
models (Fig. 1).
New technology has broadened the way in which genetic
variation can be evaluated. Traditional base by base
1505
1506
M. A. Ream and A. D. Patel
Key Points
•
•
•
Table 1. Syndromes and some of the more commonly
associated genes
Syndrome
Proper application of genetic testing in pediatric epilepsy requires understanding of the advantages and
limitations of different testing modalities
Genetic testing can be expensive, but potentially helpful
Genetic counseling is helpful in consenting for testing
and result interpretation
sequencing (Sanger sequencing) is time consuming and targets one gene at a time but is highly accurate and generally
less expensive per test ordered than newer alternatives. Next
generation, or massively parallel, sequencing (NGS) allows
for rapid sequencing of large numbers of DNA segments
that are broken into smaller pieces, sequenced, and then realigned and analyzed computationally. NGS has made large
gene panels, whole exome sequencing (WES), and even
whole genome sequencing (WGS) possible. Gene panels
sequence a list of genes known to be associated with a specific phenotype (e.g., X-linked intellectual disability, infantile
onset epilepsy, etc.). As research increases our understanding, new genes are often added to the list.
WES offers a broad evaluation for genetic variation by
sequencing most of the protein-encoding exons and splice
junctions in a patient’s genome, as it is not limited to a list of
genes known to be associated with a phenotype. However,
costs for techniques using NGS are usually high due to the
labor intensive interpretation of the vast amount of data that
is collected; but per gene sequenced, NGS is cheaper than
Sanger sequencing. The data obtained from NGS often
includes many variant of unclear significance (VUS) that
require interpretation but may not aid in a diagnosis. Parental
samples are often required to further classify a VUS. Whole
genome sequencing evaluates most of the DNA content of
the entire genome but is not available clinically at this time.
NGS does not provide a panacea for genetic diagnosis.
Mutations in noncoding areas and introns are not covered by
NGS technology as applied to WES. Triplet repeats, as in
fragile X, abnormal methylation, as in Angelman syndrome,
and some large insertions, deletions, and duplications can
be missed by WES. In general turn-around time for NGS is
also much longer (up to 4 months) than for single gene
sequencing due to the increased resources needed to interpret the results.
In addition to technical difficulties, genetic testing is not
without its controversy and ethical dilemmas.4 Confusion
can exist in the results obtained and their implications to a
patient. Ethical considerations such as paternity and discovery of genes related to diseases with later onset, such as Alzheimer’s, must be considered for WES. Once a genetic
diagnosis is obtained, the family may subsequently carry a
“label” and other family members may unwillingly become
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
Benign familial
neonatal convulsions
Benign familial infantile
convulsions
X-linked infantile spasms
Glucose transporter
deficiency syndrome
Pyridoxine dependent
seizures
Pyridoxal phosphate
responsive seizures
Folate deficiency
Dravet syndrome
GEFS+
EFMR
PME
Focal epilepsies
Epileptic
encephalopathies
Genes
KCNQ2
KCNQ3
PRRT2
SCN2A
CDKL5
ARX
SLC2A
ALDH1A7
PNPO
FOLR
SCN1A
SCN2A
GABRG2
GABRA1
PCDH19
STXBP1
HCN1
SCN1A
SCN2A
SCN1B
GABA-A
GABRD
GABRG2
PCDH19
EMP2A/laforin
EMP2B/malin
cystatin B
CHRNA4
CHRNA2
CHRNB2
Multiple
Comments
Testing provides definitive
diagnosis
Testing provides definitive
diagnosis
Testing provides definitive
diagnosis
Testing leads to definitive
treatment
Testing leads to definitive
treatment
Testing leads to definitive
treatment
Testing leads to definitive
treatment
Low MTHFR also seen in
3-phosphoglycerate
dehydrogenase deficiency
and some mitochondrial
disorders
Testing provides diagnosis
with treatment implications
Testing provides
definitive diagnosis
Testing provides definitive
diagnosis
Testing provides definitive
diagnosis with some
treatment implications
Testing provides definitive
diagnosis
Testing proves definitive
diagnosis and allows for
prognosis
This list is not exhaustive and is frequently added to by new literature.
aware of their potential carrier status. Families should meet
with a genetic counselor prior to ordering WES and often
before ordering large panels. Genetic counseling is also
usually required once results are available to help interpret
and apply results to specific patients. Many testing companies also have genetic counselors available to discuss results
with ordering providers. It is important to note that genetic
testing is currently expensive and is not covered by all insurance plans, thus potentially leading to significant out of
pocket expenses for patients, their families, and institutions.
Because of the complexity of genetic testing, many tests
require lengthy consent forms that take time and expertise
1507
Genetic Testing in Pediatric Epilepsy
Figure 1.
Classification of genetic variants. Variants are either benign or pathogenic, but determining into which category a previously undescribed
variant falls can be complicated. Variants that do not change protein structure generally would be predicted to be benign. Variants that
likely result in a change in protein structure is usually predicted to be pathogenic (for example, a premature stop codon or substitution of
a hydrophobic amino acid in a transmembrane domain of a protein for a hydrophilic amino acid). Prediction modeling also considers conservation of specific amino acid sequences across species. A highly conserved residue (i.e., the same in multiple levels of organisms) is presumed to be important for protein function so that a variant changing a highly conserved amino acid is more likely to be pathogenic.
Parental sequencing is also a useful tool in determining the significance of a VUS. If an affected child carries a de novo mutation in a gene
that can be related to the patient’s phenotype, the variant is more likely to be pathologic than if an unaffected parent carried the same variant. Sometimes prediction programs offer conflicting conclusions and they can have suboptimal sensitivity and specificity in their prediction for the impact of a variant,80 leaving the final interpretation of the significance of a result up to the clinician.
Epilepsia ILAE
to properly complete and can be cumbersome for both families and providers.
The wide availability of large-scale genetic testing brings
an overlap between clinical medicine and research since
research findings can be applied for clinical purposes. Foregoing the need for an a priori genetic diagnosis, broad genetic
screening such as with WES allows gene discovery. A VUS
identified in one patient can be reclassified as disease causing
if similar associations are found in other patients. Clinical
discovery of variants can inform basic research regarding
mechanisms that provide therapeutic targets.5 In addition,
simple clinical observation regarding therapeutic responses
in specific conditions can lead to further discovery of
involved mechanisms, such as was noted in the exacerbating
effect of sodium channel blockers in Dravet syndrome.6
Despite the practical and theoretical challenges of genetic
testing, genetic testing may offer advantages over retaining
a diagnosis of “idiopathic” epilepsy. Certain results may
guide and alter treatment decisions as in Dravet syndrome.
The benefit of providing an answer to the diagnostic odyssey cannot be understated. Family anxiety can be eased, a
more specific prognosis may be available, additional at-risk
relatives can be tested, and a specific diagnosis may lead to
networking with other similarly affected families. A diagnosis and its inheritance pattern may assist in future family
planning; some families learn that they have a 25% or 50%
risk of having another affected child, whereas other families
are relieved to find out that their affected child carries a de
novo mutation.
Selected Genetic Epilepsy
Syndromes
Benign familial neonatal convulsions (BFNC)
Benign familial neonatal convulsions (BFNC) is an
autosomal dominant condition presenting with focal or
generalized seizures, sometimes with apneas or autonomic
symptoms, that generally begin between 2 and 8 days of
life. Seizures resolve in approximately 67% of patients by
6 weeks of life. Neurodevelopment is usually normal,
although there is an increased risk of subsequent epilepsy
(16%).7 Those patients who do develop epilepsy often
develop seizures suggestive of benign epilepsy with centrotemporal spikes (BECTS). Treatment with phenobarbital
is usually initiated due to the flurry with which seizures present, with up to 30 per day; however, treatment may not
affect seizure recurrence or cessation.
Genes involved in BFNC include KCNQ3 (Kv7.3)8 and
KCNQ2 (Kv7.2).9 Seventy percent of cases have a mutation
identified in one of these genes, with KCNQ2 responsible
for 90% of genetically identified cases.10 De novo and
inherited KCNQ2 mutations have similar prognosis.11
Mutations in KCNQ2 and KCNQ3 have also been reported
in patients with BECTS with and without a history of neonatal convulsions.12
KCNQ2 encephalopathy
Some patients with KCNQ2 mutations experience a much
more severe course, with epileptic encephalopathy, refractory seizures, and/or intellectual disability.13 In the first
week of life, patients present with tonic seizures and an initial burst suppression electroencephalography (EEG) pattern. Seizures are typically eventually controlled with
sodium channel blockers, but all patients have severe intellectual disability. Such outcomes have been reported to arise
from de novo mutations and from families with a history of
typical BFNC.14
Benign familial infantile convulsions (BFIC)
Benign familial infantile convulsions demonstrate more
genetic heterogeneity than BFNC. Onset is typically
between 3 and 10 months and consists of brief seizures with
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
1508
M. A. Ream and A. D. Patel
head and eye deviation, tonic stiffening, and cyanosis with
unilateral clonic limb movements that become bilateral.
Seizures are associated with occipitoparietal discharges that
generalize.15 Seizures occur in clusters over 1–4 days and
then generally subside within 1 year. Interictal EEG and
development are normal. Cases with and without family
history have a similar clinical course.16
The majority (70%) of cases of BFIC are linked to PRRT2
mutations,17 which are associated with pure BFIC, BFIC
with choreoathetosis,18 and paroxysmal kinesigenic dyskinesia. Seizures generally respond well to antiepileptic medications including carbamazepine, phenobarbital, valproate, or
zonisamide.19 However, if sodium channel blockers are used
without effect, mutation in SCN2A must be considered.20
X-linked infantile spasms (CDKL5 and ARX)
Mutations in CDKL5 are inherited in an X-linked dominant fashion and are a common cause of infantile spasms
(IS) and early onset seizures. CDKL5 is responsible for 8%
of early onset seizures (<9 months) in girls and 28% of early
onset seizures with infantile spasms in girls.21 Frequency in
boys is more difficult to ascertain due to fetal loss, but
CDKL5 mutations were found in 3% of boys undergoing
CDKL5 sequencing as part of an evaluation for early onset
intractable epilepsy.22
CDKL5 was first described in cases of atypical Rett syndrome in patients with acquired microcephaly, developmental delay, limited language, hand apraxia, and chorea, but
who had infantile onset epilepsy and severe hypotonia, and
often lacked a period of regression.23 The first family
described illustrates the phenotypic variability24; one sister
had atypical Rett syndrome. Her twin sister had autism and
mild-to-moderate intellectual disability without seizures.
Their brother, who also carried the mutation, had LennoxGastaut syndrome, profound intellectual impairment, spastic quadriparesis, and cortical blindness. He eventually
became “unresponsive” to environmental stimuli. Diagnosis
of CDKL5 mutation may help prognosis and limit further
testing but does not necessarily affect treatment.
ARX mutation is a rare cause of X-linked recessive intellectual disability,25 either as a sole symptom or in combination with a variety of other neurologic abnormalities
including infantile spasms, epilepsy, dystonia (Partington
syndrome) and autism.26 Loss of function mutations are
associated with brain malformations including lissencephaly (often with abnormal testes), hydranencephaly, and
agenesis of the corpus callosum.27 ARX de novo mutations
were found in 3 of 264 patients with epileptic encephalopathy (IS or Lennox-Gastaut syndrome),28 whereas mutation
was found in 2.8% of boys undergoing ARX sequencing as
part of an evaluation for early onset intractable epilepsy and
3% of boys with X-linked intellectual disability who were
FRX negative.29
Seizure types associated with ARX mutations include IS,
generalized tonic–clonic, myoclonic, absence seizures, and
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
complex partial. Microcephaly, macrocephaly, and
increased and decreased tone have all been associated with
mutations in boys. Intellectual impairment can be mild to
severe, with more severe forms often associated with IS.30
Most reported cases are male, but female patients with
ARX mutations can have learning and cognitive disabilities,
autism, neuropsychiatric disorders (anxiety, depression,
and schizophrenia), epilepsy, and agenesis of the corpus
callosum. The severe phenotype seen in boys with intellectual impairment and IS can also be seen in girls if there is
preferential X inactivation of the normal allele.31 ARX is
considered an interneuronopathy resulting in loss of
c-aminobutyric acid (GABA)ergic interneurons.32 This
observation may eventually help in selecting specific antiepileptic drugs, for example, vigabatrin, which enhances
GABA signaling, but studies on pharmacogenetics are still
lacking.
Glucose 1 transporter deficiency
Glucose transporter deficiency syndrome type 1 (GLUTDS) is an autosomal dominant condition due to mutation in
the glucose transporter and presents with seizures in the first
year of life with acquired microcephaly, spasticity, and
ataxia. Seizure types in infants include atypical absences
and infantile spasms, and in older children generalized
tonic–clonic seizures and typical absence.33 Other less common phenotypes include ataxia, exercise-induced dyskinesia, intellectual disability without seizures, and alternating
hemiplegia of childhood.34 GLUT-DS has been associated
with 10% of patients with early onset absence epilepsy and
1% of idiopathic generalized epilepsy. 35
In the classical phenotype, diagnosis can be made based
on low cerebrospinal fluid (CSF) glucose content
(<2.5 mmol/l) and low CSF to serum glucose ratio (generally <0.5)36; however, the sensitivity of CSF analysis has
been questioned.37 Diagnosis can be confirmed by sequencing the SLC2A gene. Treatment with the ketogenic diet
improves seizures and movement disorders. Steroids and
carbonic anhydrase inhibitors can also be helpful.38
Pyridoxine-dependent epilepsy
Pyridoxine-dependent epilepsy is caused by recessive
mutations in ALDH1A7, the gene coding antiquitin, which
is an a-amino adipic semialdehyde (AASA) dehydrogenase
that is part of the lysine catabolism pathway. Deficiency
will result in defective conversion of glutamate to GABA,
leading to excess neuroexcitation.39 The abnormal metabolism can also cause structural brain defects that are not corrected by postnatal B6 supplementation.40 The classic form
causes medication-resistant seizures in the first few hours of
life,41 although atypical presentations have been described
with later onset seizures and seizure-free periods without
pyridoxine supplementation. Diagnosis can be suggested by
elevated AASA or pipecolic acid in urine, plasma, and CSF
whereas urine organic acid profile shows elevated lysine. A
1509
Genetic Testing in Pediatric Epilepsy
lysine-restricted diet may be helpful in preventing the toxicity of accumulated metabolites and reduce seizure frequency.42 The most satisfying diagnostic method is
observation of clinical and electrographic improvement
within minutes of administering parenteral pyridoxine,43
although the short-term response is not always diagnostic.
Definitive diagnosis is through CSF measurement and gene
sequencing.
such as carbamazepine and phenytoin has been associated
with seizure aggravation and is contraindicated. A de novo
genetic variation in the SCN1A gene occurs in approximately 80% of patients, mostly due to single nucleotide
mutations or small insertions or deletions.52 Other genes
recently associated with the Dravet phenotype include
SCN2A, GABRG2, GABRA1, PCDH19, STXBP1, and
HCN1.53,54
Pyridoxal phosphate–dependent epilepsy
Mutation in PNPO, which encodes pyridoxine 50 -phosphate (P5P) oxidase, prevents conversion of pyridoxine to
its biologically active form and results in P5P-dependent
epilepsy. Seizures can start in utero, within hours of birth, or
as late as 2 years of age.44 Seizures may present with a variety of characteristics including myoclonic, tonic, clonic,
tonic status epilepticus, neonatal generalized tonic–clonic
seizures, and electrical status epilepticus of sleep (ESES).44
Biochemical findings include elevated glycine and threonine on the plasma amino acid profile, elevated CSF
L-DOPA and 3-methoxytyrosine, and reduced urine
homovanillic acid and 5-hydroxyindoleacetic acid.45 Treatment with pyridoxal-5-phosphate 30–50 mg/kg/day divided
in three to four doses can lead to reduction in seizure recurrence and improvement of cognitive functions.46
Generalized epilepsy febrile seizures plus (GEFS+)
GEFS+ refers to an autosomal dominant epilepsy syndrome with incomplete penetrance. Patients present with
generalized tonic–clonic seizures often associated with
fever, usually presenting or continuing beyond 6 years of
age when simple febrile seizures abate. They may also present with afebrile epileptic seizures. Other seizure types can
be present as well. GEFS+ is associated with a variation in
the SCN1A gene, which encodes the a1 subunit of a voltage
gated sodium channel.55 Other genes associated with
GEFS+ include SCN2A, SCN1B, GABA-A, GABRD, and
GABRG2.56 Testing for these patients may not be helpful as
a full-treatment response, and remission often occurs by the
preteen years.57
Cerebral folate deficiency
Cerebral folate deficiency is due to either autoimmunity
against or mutation in FOLR and can be detected as low
CSF methyl tetrahydrofolate (MTHF). Folate is converted
to 5MTHF, which is used for production of purines and
recycling of homocysteine into s-adenosylmethionine.
Autoimmune cerebral folate deficiency presents as sleep
disturbance and irritability in infancy and can progress to
seizures, autism, dyskinesia, ataxia, and spasticity if
untreated.47 Folate deficiency caused by receptor gene mutation presents in childhood as neurodegeneration with
seizures, developmental delay, ataxia, hypotonia, and, in
adolescents, severe polyneuropathy. Low MTHF can also be
seen in 3-phosphoglycerate dehydrogenase deficiency (a congenital serine biosynthesis disorder), and mitochondrial disorders such as Kearns-Sayre syndrome or Alpers disease.48
Treatment of cerebral folate deficiency, either due to
autoimmune or receptor mutation, is with folinic acid (0.5–
1 mg/kg/day).49 Folic acid is contraindicated for these
patients, as folic acid competes at FR1 with the biologically
active form MTHF and can worsen symptoms.50
Dravet syndrome (DS)
Children with DS often present with generalized febrile
or hemiconvulsive seizures that start during the first year of
life.51 Eventually, myoclonic and other seizure types occur.
As the seizures continue and often become treatment
resistant, developmental slowing or regression occurs.
Treatment with sodium channel blocking anticonvulsants
Epilepsy limited to females with mental retardation
(EFMR)
Girls with epilepsy limited to females with mental retardation (EFMR) present in infancy with focal or generalized
tonic–clonic febrile seizures, with clustering occurring
within a few days of the first seizure. Often, these seizures
are difficult to control with treatment. At seizure onset,
developmental plateauing or regression occurs.58 Many of
the girls have autistic features. A variation in PCDH19
encoding protocadherin 19 is felt to be associated with
EFMR, with a unique mode of X-linked inheritance in
which only heterozygous females and mosaic males are
affected.59 Overlap with symptoms seen in Dravet syndrome can complicate the diagnosis. Therefore, PCDH19
testing should be considered in female patients with a Dravet-like phenotype or in those females with previous negative SCN1A testing.
Progressive myoclonic epilepsy (PME)
Adolescent-onset myoclonic epilepsy most often represents juvenile myoclonic epilepsy, but a small subset of
these patients will later prove to have PME-targeted
sequencing of EMP2A/laforin and EMP2B/malin (which
cause Lafora progressive myoclonic epilepsy), and cystatin
B (Unverricht-Lundborg disease) can distinguish these
catastrophic diseases from juvenile myoclonic epilepsy in
the early stages in which they may not be clinically distinguishable.60 A diagnosis of Lafora disease may lead to targeted therapy with a premature stop codon read-through
drug (such as gentamicin) and avoidance of drugs that can
exacerbate myoclonia and dementia such as sodium channel
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
1510
M. A. Ream and A. D. Patel
blockers and GABAergic drugs.61 Such testing can be low
yield and expensive and should be reserved for cases that
are resistant to treatment. There are also panels with genes
related to myoclonic epilepsy. A patient’s insurance coverage may help determine if a panel or multiple targeted genes
done in series is more cost-effective.
Genetic focal epilepsies
There has been significant advancement recently in the
genetics of focal epilepsies.62 Variation in several acetylcholine receptors led to autosomal dominant familial focal
seizures, including nocturnal frontal lobe epilepsy and temporal lobe epilepsy.63,64 Focal epilepsy due to brain structural abnormalities also often has a genetic etiology as in
DEPDC5.65 There is wide phenotypic variation in genes
associated with focal epilepsies, as seen in KCNT1 and
GRIN2A, both of which can cause focal epilepsy and epileptic encephalopathy.63,66,67 Such variability makes broad
genetic evaluation much more practical than single gene
sequencing in most cases.
seizures.69 As new genes and associations are discovered,
the indications for genetic testing may change in the future.
For drug-resistant epilepsy, unless a specific disorder is
highly suspected in the situations discussed in subsequent
text, we recommend the diagnostic approach described in
Figure 2. Initial biochemical screening should be done,
when indicated, to uncover treatable metabolic disorders
quickly while awaiting genetic confirmation. Simultaneously, genetic evaluation can be initiated with microarray
followed by gene panel and then WES if diagnosis is still
sought.
Direct biochemical testing
Seizures can be the presenting symptom of inborn errors
of metabolism (IEMs) or can occur later in the course of dis-
Epileptic encephalopathies
Many of the above-listed genes and many additional
genes (for example SCN8A, GNAO1, CHD2, SIK1, SYNGAP1, SLC6A1, KCNA2, and DNM1) can result in epileptic
encephalopathy. Epilepsy panels cover the better-described
genes. However, it is clear from exome-sequencing studies
that many genes are potential candidates and many of these
candidates fall within the same functional network.28,68
Computational biology will help us further delineate genes
and gene networks involved in etiology and treatment of
epilepsy as science advances.
Proposed Algorithm for
Diagnosis of Drug-Resistant
Epilepsy
If a genetic disorder is suspected as the cause of epilepsy,
based on drug resistance, family history, or clinical phenotype, timely diagnosis may reduce overall cost, limit the “diagnostic odyssey” that can bring much anxiety to families,
improve prognostication, and lead to targeted therapy.
Recent ILAE recommendations indicate that genetic counseling should be available to all epilepsy patients, but that
genetic evaluation should be undertaken at a tertiary level of
epilepsy care, to which infants with seizure and patients of
any age after failure of one antiepileptic drug should be
referred.69 Generally, genetic testing is not recommended in
drug-responsive epilepsy or at epilepsy onset, although comparative genomic hybridization (CGH) can be used as firsttier evaluation of patients with global developmental delay,
which is a population that is at higher risk of epilepsy. Metabolic testing should be undertaken at the onset of epilepsy in
infants without structural or syndromic cause of their
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
Figure 2.
Algorithm for the diagnostic evaluation of drug resistant epilepsy.
Epilepsia ILAE
1511
Genetic Testing in Pediatric Epilepsy
ease. IEMs should be suspected as a potential cause of
epilepsy in infants, particularly if the clinical history is not
consistent with hypoxic ischemic encephalopathy or if there
were in utero seizures. Other clinical scenarios for which to
consider IEM include cases of myoclonic epilepsy, infantile
spasms, atypical absences, epilepsia partialis continua, episodic decompensation, and when hypsarrhythmia or burst
suppression is present on EEG (Table 2). In addition, by the
time a patient has established drug-resistant epilepsy, they
should have already undergone magnetic resonance imaging (MRI) to rule out structural lesions as the cause of epilepsy; such imaging may provide clues to IEMs that may
coexist with brain lesions.70 All patients with suspected
IEM should have initial metabolic screening (Table 3).
Additional studies may be performed depending on the clinical scenario. Several authors have provided reviews to
direct such testing in specific circumstances.71
Results can be obtained from metabolic evaluation much
faster than from either targeted genetic studies or panels that
include genes associated with these disorders, allowing for
rapid initiation of specific treatments, when available.
Genetic confirmation is helpful when there is a biochemical
diagnosis and allows noninvasive screening of other family
members if needed.
Single gene sequencing
Earlier we discussed a list of genetic epilepsies, but most
have a variable phenotype, making identification of “classical phenotypes” less likely. However, there are a few condi-
Table 2. Clinical situations in which to suspect an inborn
error of metabolism.
Neonates with history of in utero seizures
Myoclonic epilepsy in infants
Infantile spasms
Atypical absence
Epilepsia partialis continua
Episodic decompensation
Hypsarrhythmia
Burst suppression
MRI with metabolic pattern
Table 3. Recommended initial metabolic screening
Electrolytes with glucose
complete blood count (CBC)
Hepatic enzymes
Plasma amino acids
Urine organic acids
Ammonia
Lactate
Acylcarnitine profile
Consider cerebrospinal fluid (CSF) analysis
If hypotonic: creatine phosphokinase (CK)
tions in which targeted sequencing may be prudent due to
issues related to treatment or prognosis and these include
the following: Dravet syndrome, paternal transmission of
early onset epilepsy and developmental delay in females,
early onset absence epilepsy, and in some situations regarding pharmacologic management (Table 4). There are several online resources for identifying classical phenotypes
and determining which commercial labs perform specific
gene sequencing including GeneReviews (http://www.ncbi.
nlm.nih.gov/books/NBK1116/) and genetests.org. Deprez
et al. suggested an algorithm for targeted genetic evaluation
of epilepsy occurring in the first year of life, but given the
interval development of gene panels, his proposed approach
that often requires sequencing of more than one gene for a
given phenotype would be more costly than a gene panel,
except perhaps in the conditions listed in Table 4.
Chromosomal analysis
Drug-resistant epilepsy, especially when associated with
developmental delay or congenital anomalies, should
prompt comparative genomic hybridization analysis (chromosomal microarray), which detects CNV. Balanced rearrangements and point mutations will not be detected, thus,
limiting utility for diseases caused by single genes.72 CGH
in patients with epilepsy and developmental delay has a
23.5% rate of abnormal findings.73 Yield increases when
multiple abnormalities are present. Costs to obtain this testing have decreased in the past several years with many, but
not all, insurance carriers providing reimbursement. Reflex
to other forms of genetic testing, such as gene panels, when
results are negative for CGH are available through some
commercial labs.
Karyotype can detect aneuploidy and balanced translocations not detectable by CGH with a resolution down to
3 Mb. A yield of 14% in epilepsy has been reported,
although numbers were small.2 Several chromosomal disorders are associated with epilepsy, warranting karyotype if
microarray is negative, and large genomic abnormality is
expected based on phenotype.
Gene panels
Most cases of drug-resistant epilepsy are not so straight
forward as to indicate a clear monogenetic diagnosis. Given
variable expressivity, incomplete penetrance, multigeneic
interactions, and de novo mutations lacking an informative
family history, targeted single gene sequencing is not practical in most cases. In addition, the ILAE reported multiple
genes for which a genetic diagnosis is considered very useful.74 If such diagnoses are not to be missed, casting a broad
net with genetic testing can be justified.
Utilizing NGS, panels were developed to evaluate genetic
variation within multiple disease-related genes simultaneously. The quality of the testing relies on the coverage available at each base-pair position on the genes being tested.75
This becomes important, as depth of coverage varies
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
1512
M. A. Ream and A. D. Patel
Table 4. Clinical situations that may warrant targeted gene sequencing
Targeted gene
sequencing
SCN1A
PCDH19
SLC2A1
POLG
HLA-B*1502
P450 and other
liver enzymes
Clinical condition
Advantage of testing
Dravet syndrome. Consider testing if recurrent episodes of febrile status
epilepticus, intractable tonic–clonic seizures during the first year of life,
epileptic encephalopathy attributed to vaccination, and in adults with a
history consistent with Dravet syndrome
Females presenting with multiple clusters of brief febrile seizures and
developmental delay or regression, particularly if there is a family
history consistent with paternal transmission.
Onset of absence seizures <4 years old, particularly if there is a
family history of paroxysmal exercise-induced dyskinesia
Prior to starting valproic acid in patients with drug resistant
seizures and developmental delay or regression
Prior to starting carbamazepine, oxcarbazepine, phenytoin and
lamotrigine in patients of Asian descent
Unexpected antiepileptic drug toxicity or unexpectedly high/low
drug levels. (still under research, and not broadly recommended)
between commercial labs, which employ different quality
measures. Coverage of the genes may be equal to or greater
than what is performed via WES. The number of genes
available on a panel range from 20 to >400, depending on
which company’s testing is utilized.
Panels broadly related to a patient’s phenotype (e.g.,
myoclonic epilepsy, infantile onset epilepsy) may be the
most cost–effective approach to genetic diagnosis because
it limits the search to known epilepsy genes but covers multiple genes at once. When a panel of 265 genes was applied
to 33 patients with refractory epilepsy, 16 received a specific diagnosis.75 Multiple gene panels of various sizes are
available commercially. Companies tout their utility, with
up to 22% sensitivity of one panel with 51 genes related to
infantile onset epilepsy (GeneDx.com).
Whole exome sequencing
Currently the broadest net one could cast with clinical
genetic evaluation is with WES because it allows simultaneous interrogation of all genes without limitation to known
epilepsy genes and includes all exons in the genome.76 The
American College of Medical Genetics recommends that
exome sequencing be considered when there is a suspected
genetic disease without a specific test available, in disorders
with genetic heterogeneity that would require sequencing of
multiple genes simultaneously for proper diagnosis, when
there is a suspected genetic disorder but specific genetic
tests are negative, and in some cases for prenatal testing
(Table 5).77 WES has proven to be helpful in the diagnosis
Table 5. Indications for WES
Suspected genetic disease without a specific test available
Disorders with genetic heterogeneity that would require sequencing of
multiple genes simultaneously
Genetic disorder is suspected, but specific genetic tests are negative
Special prenatal circumstances
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
Avoidance of sodium channel blockers, aggressive
seizure management, justification of stiripentol,
bromides, etc.
Prognosis and potential forthcoming treatment options
Initiation of ketogenic diet
Avoidance of potentially fatal liver failure starting as
early as 2 months after initiation of valproic acid therapy
Avoidance of potentially fatal reaction (Stevens-Johnson
syndrome/toxic epidermal necrolysis)
Avoidance of drugs metabolized through an aberrant
P450 system
of epilepsy. Protein-altering mutations in genes related to,
or potentially related to, epilepsy were found in 9 of 10
patients with epileptic encephalopathy undergoing WES.68
Of 264 patients with epileptic encephalopathies, 329 de
novo mutations were found, whereas only a fraction of these
were known to be pathogenic. Such studies highlight the
complexity of interpreting WES data and the potential for
gene discovery if multiple patients with similar phenotypes
carry similar variants or if variants occur in a common biochemical pathway.28 Commercially, 39% sensitivity for
WES applied to epilepsy trios has been advertised
(GeneDx.com). Definitive diagnostic yield has been
reported as high as 45% in patients with neurodevelopmental disabilities, and these authors argue that WES followed
by whole genome sequencing can be cost-effective and even
less expensive than traditional molecular diagnostic
approaches.78 In common application, WES results may
vary; one retrospective study reported 14.3% diagnosis
based on WES in treatment-resistant pediatric epilepsy,2
and another reported 27% yield for patients with a variety of
neurodevelopmental conditions, including epilepsy.79
For the most useful interpretation of a patient’s WES, it is
preferable to also obtain testing from both biologic parents
to allow identification of de novo variants, which are more
likely to be disease causing than variants inherited from
unaffected parents. Although WES gives the largest volume
of data regarding a patient’s genetics and requires no prior
diagnosis, it has several limitations—including cost, turnaround-time, ethical concerns, and the potential for irrelevant or unwanted data—and should not be considered a
first-line test. Reanalysis of data is sometimes also necessary. As more data are reported in the literature, some variants previously classified as of unknown significance may
be reclassified as pathogenic. Thus, continuing to publish
novel findings in the medical literature continues to be
important and considering periodic repeat analysis of previously reported VUS may be beneficial.
1513
Genetic Testing in Pediatric Epilepsy
Costs for WES range from $5,000 to >$14,000, with most
insurance carriers not providing reimbursement as they consider such testing experimental, although institutional and
bulk discounts may sometimes apply. Precertification is recommended and often requires multiple steps of appeals to
obtain authorization. If metabolic testing, microarray, and
gene panel are not informative but a genetic diagnosis is still
sought, the physician and family must weigh the many pros
and cons of pursing further genetic testing with WES.
Summary
The use of genetic testing in pediatric epilepsy is complicated and the list of known epilepsy genes changes almost
daily. Also as reimbursement for testing changes, the most
cost-effective approach may change over time. Currently
we recommend reserving WES for the most elusive cases,
but in time sending one genetic test with the broadest
genetic coverage may become practical; however, there are
still technical limitations that will need to be addressed as
WES becomes more utilized. It would be ideal if WES could
be tailored for epilepsy patients by optimizing coverage of
known epilepsy genes. If this were to occur, further need for
testing with a panel first would not be necessary and would
ultimately lead to cost savings by reducing the total number
of tests ordered. Alternatively, reflex testing to WES is an
option when an epilepsy gene panel is negative. A few commercial companies currently offer this option.
Careful consideration of each available test with its limitations must precede selection and ordering. A close partnership with genetics and genetic counseling is strongly
recommend before pursuing testing that may reveal unexpected, and sometimes unrequested, results. Interpretation
of complicated results should be performed only in collaboration with genetics providers unless the ordering neurologist has a strong background in and understanding of
genetics. Genetic testing can play an important role in the
care provided to patients with epilepsy.
Disclosures
Neither of the authors has any conflicts of interest to disclose that are relevant to this manuscript. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is
consistent with those guidelines.
References
1. Cowan LD, Bodensteiner JB, Leviton A, et al. Prevalence of the
epilepsies in children and adolescents. Epilepsia 1989;30:94–106.
2. Ream MA, Mikati MA. Clinical utility of genetic testing in pediatric
drug-resistant epilepsy: a pilot study. Epilepsy Behav 2014;37:241–
248.
3. Berg AT, Berkovic SF, Brodie MJ, et al. Revised terminology and concepts for organization of seizures and epilepsies: report of the ILAE
Commission on Classification and Terminology, 2005–2009. Epilepsia
2010;51:676–685.
4. Pal DK, Pong AW, Chung WK. Genetic evaluation and counseling for
epilepsy. Nat Rev Neurol 2010;6:445–453.
5. Thomas RH, Berkovic SF. The hidden genetics of epilepsy-a clinically
important new paradigm. Nat Rev Neurol 2014;10:283–292.
6. Millichap JJ, Koh S, Laux LC, et al. Child neurology: Dravet syndrome: when to suspect the diagnosis. Neurology 2009;73:e59–e62.
7. Shevell MI, Sinclair DB, Metrakos K. Benign familial neonatal seizures: clinical and electroencephalographic characteristics. Pediatr
Neurol 1986;2:272–275.
8. Charlier C, Singh NA, Ryan SG, et al. A pore mutation in a novel
KQT-like potassium channel gene in an idiopathic epilepsy family. Nat
Genet 1998;18:53–55.
9. Singh NA, Charlier C, Stauffer D, et al. A novel potassium channel
gene, KCNQ2, is mutated in an inherited epilepsy of newborns. Nat
Genet 1998;18:25–29.
10. Soldovieri MV, Miceli F, Bellini G, et al. Correlating the clinical and
genetic features of benign familial neonatal seizures (BFNS) with the
functional consequences of underlying mutations. Channels (Austin)
2007;1:228–233.
11. Claes LR, Ceulemans B, Audenaert D, et al. De novo KCNQ2 mutations in patients with benign neonatal seizures. Neurology
2004;63:2155–2158.
12. Neubauer BA, Waldegger S, Heinzinger J, et al. KCNQ2 and KCNQ3
mutations contribute to different idiopathic epilepsy syndromes. Neurology 2008;71:177–183.
13. Dedek K, Fusco L, Teloy N, et al. Neonatal convulsions and epileptic
encephalopathy in an Italian family with a missense mutation in
the fifth transmembrane region of KCNQ2. Epilepsy Res 2003;54:21–
27.
14. Borgatti R, Zucca C, Cavallini A, et al. A novel mutation in KCNQ2
associated with BFNC, drug resistant epilepsy, and mental retardation.
Neurology 2004;63:57–65.
15. Vigevano F, Fusco L, Di Capua M, et al. Benign infantile familial convulsions. Eur J Pediatr 1992;151:608–612.
16. Caraballo RH, Cersosimo RO, Espeche A, et al. Benign familial and
non-familial infantile seizures: a study of 64 patients. Epileptic Disord
2003;5:45–49.
17. Zara F, Specchio N, Striano P, et al. Genetic testing in benign familial
epilepsies of the first year of life: clinical and diagnostic significance.
Epilepsia 2013;54:425–436.
18. Caraballo R, Pavek S, Lemainque A, et al. Linkage of benign familial
infantile convulsions to chromosome 16p12-q12 suggests allelism to
the infantile convulsions and choreoathetosis syndrome. Am J Hum
Genet 2001;68:788–794.
19. Vigevano F. Benign familial infantile seizures. Brain Dev
2005;27:172–177.
20. Kwan P, Poon WS, Ng HK, et al. Multidrug resistance in epilepsy and
polymorphisms in the voltage-gated sodium channel genes SCN1A,
SCN2A, and SCN3A: correlation among phenotype, genotype, and
mRNA expression. Pharmacogenet Genomics 2008;18:989–998.
21. Nemos C, Lambert L, Giuliano F, et al. Mutational spectrum of
CDKL5 in early-onset encephalopathies: a study of a large collection
of French patients and review of the literature. Clin Genet
2009;76:357–371.
22. Mirzaa GM, Paciorkowski AR, Marsh ED, et al. CDKL5 and ARX
mutations in males with early-onset epilepsy. Pediatr Neurol
2013;48:367–377.
23. Bahi-Buisson N, Nectoux J, Rosas-Vargas H, et al. Key clinical features to identify girls with CDKL5 mutations. Brain 2008;131:2647–
2661.
24. Weaving LS, Christodoulou J, Williamson SL, et al. Mutations of
CDKL5 cause a severe neurodevelopmental disorder with infantile
spasms and mental retardation. Am J Hum Genet 2004;75:1079–1093.
25. Grønskov K, Hjalgrim H, Nielsen IM, et al. Screening of the ARX
gene in 682 retarded males. Eur J Hum Genet 2004;12:701–705.
26. Suri M. The phenotypic spectrum of ARX mutations. Dev Med Child
Neurol 2005;47:133–137.
27. Kato M, Das S, Petras K, et al. Mutations of ARX are associated with
striking pleiotropy and consistent genotype–phenotype correlation.
Hum Mutat 2004;23:147–159.
28. Allen AS, Berkovic SF, Cossette P, et al. De novo mutations in epileptic encephalopathies. Nature 2013;501:217–221.
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
1514
M. A. Ream and A. D. Patel
29. Nawara M, Szczaluba K, Poirier K, et al. The ARX mutations: a frequent cause of X-linked mental retardation. Am J Med Genet A
2006;140:727–732.
30. Scheffer IE, Wallace RH, Phillips FL, et al. X-linked myoclonic epilepsy with spasticity and intellectual disability: mutation in the homeobox gene ARX. Neurology 2002;59:348–356.
31. Kalscheuer VM, Tao J, Donnelly A, et al. Disruption of the serine/threonine kinase 9 gene causes severe X-linked infantile spasms and mental retardation. Am J Hum Genet 2003;72:1401–1411.
32. Kato M, Dobyns WB. X-linked lissencephaly with abnormal genitalia
as a tangential migration disorder causing intractable epilepsy: proposal for a new term, “interneuronopathy”. J Child Neurol
2005;20:392–397.
33. Vykuntaraju KN, Bhat S, Sanjay KS, et al. Symptomatic West syndrome secondary to glucose transporter-1(GLUT1) deficiency with
complete response to 4:1 ketogenic diet. Indian J Pediatr
2014;81:934–936.
34. Tzadok M, Nissenkorn A, Porper K, et al. The many faces of Glut1
deficiency syndrome. J Child Neurol 2014;29:349–359.
35. De Giorgis V, Veggiotti P. GLUT1 deficiency syndrome 2013: current
state of the art. Seizure 2013;22:803–811.
36. Leen WG, Klepper J, Verbeek MM, et al. Glucose transporter-1 deficiency syndrome: the expanding clinical and genetic spectrum of a
treatable disorder. Brain 2010;133:655–670.
37. Mullen SA, Suls A, De Jonghe P, et al. Absence epilepsies with widely
variable onset are a key feature of familial GLUT1 deficiency. Neurology 2010;75:432–440.
38. Vieker S, Schmitt J, Langler A, et al. Unusual sensitivity to steroid
treatment in intractable childhood epilepsy suggests GLUT1 deficiency syndrome. Neuropediatrics 2012;43:275–278.
39. Baumeister FA, Gsell W, Shin YS, et al. Glutamate in pyridoxine-dependent epilepsy: neurotoxic glutamate concentration in the cerebrospinal fluid and its normalization by pyridoxine. Pediatrics
1994;94:318–321.
40. Jansen LA, Hevner RF, Roden WH, et al. Glial localization of antiquitin: implications for pyridoxine-dependent epilepsy. Ann Neurol
2014;75:22–32.
41. Hunt AD, Stokes J, McCRORY WW, et al. Pyridoxine dependency:
report of a case of intractable convulsions in an infant controlled by
pyridoxine. Pediatrics 1954;13:140–145.
42. van Karnebeek CD, Hartmann H, Jaggumantri S, et al. Lysine
restricted diet for pyridoxine-dependent epilepsy: first evidence and
future trials. Mol Genet Metab 2012;107:335–344.
43. Mikati MA, Trevathan E, Krishnamoorthy KS, et al. Pyridoxine-dependent epilepsy: EEG investigations and long-term follow-up. Electroencephalogr Clin Neurophysiol 1991;78:215–221.
44. Veerapandiyan A, Winchester SA, Gallentine WB, et al. Electroencephalographic and seizure manifestations of pyridoxal 50 -phosphatedependent epilepsy. Epilepsy Behav 2011;20:494–501.
45. Bagci S, Zschocke J, Hoffmann GF, et al. Pyridoxal phosphate-dependent neonatal epileptic encephalopathy. Arch Dis Child Fetal Neonatal
Ed 2008;93:F151–F152.
46. Hoffmann GF, Schmitt B, Windfuhr M, et al. Pyridoxal 50 -phosphate
may be curative in early-onset epileptic encephalopathy. J Inherit
Metab Dis 2007;30:96–99.
47. Hyland K, Shoffner J, Heales SJ. Cerebral folate deficiency. J Inherit
Metab Dis 2010;33:563–570.
48. Garcia-Cazorla A, Quadros EV, Nascimento A, et al. Mitochondrial
diseases associated with cerebral folate deficiency. Neurology
2008;70:1360–1362.
49. Surtees R, Wolf N. Treatable neonatal epilepsy. Arch Dis Child
2007;92:659–661.
50. Wu D, Pardridge WM. Blood–brain barrier transport of reduced folic
acid. Pharm Res 1999;16:415–419.
51. Dravet C. Dravet syndrome history. Dev Med Child Neurol 2011;53
(Suppl. 2):1–6.
52. Mullen SA, Scheffer IE. Translational research in epilepsy genetics:
sodium channels in man to interneuronopathy in mouse. Arch Neurol
2009;66:21–26.
53. Nava C, Dalle C, Rastetter A, et al. De novo mutations in HCN1
cause early infantile epileptic encephalopathy. Nat Genet 2014;46:
640–645.
Epilepsia, 56(10):1505–1514, 2015
doi: 10.1111/epi.13122
54. Depienne C, Trouillard O, Bouteiller D, et al. Mutations and deletions
in PCDH19 account for various familial or isolated epilepsies in
females. Hum Mutat 2011;32:E1959–E1975.
55. Mulley JC, Scheffer IE, Petrou S, et al. SCN1A mutations and epilepsy. Hum Mutat 2005;25:535–542.
56. Baulac S, Huberfeld G, Gourfinkel-An I, et al. First genetic evidence
of GABA(A) receptor dysfunction in epilepsy: a mutation in the gamma2-subunit gene. Nat Genet 2001;28:46–48.
57. Scheffer IE, Berkovic SF. Generalized epilepsy with febrile seizures
plus. A genetic disorder with heterogeneous clinical phenotypes. Brain
1997;120(Pt 3):479–490.
58. Camacho A, Simon R, Sanz R, et al. Cognitive and behavioral profile
in females with epilepsy with PDCH19 mutation: two novel mutations
and review of the literature. Epilepsy Behav 2012;24:134–137.
59. Depienne C, Bouteiller D, Keren B, et al. Sporadic infantile epileptic
encephalopathy caused by mutations in PCDH19 resembles Dravet
syndrome but mainly affects females. PLoS Genet 2009;5:e1000381.
60. Delgado-Escueta AV, Bourgeois BF. Debate: Does genetic information in humans help us treat patients? PRO–genetic information in
humans helps us treat patients. CON–genetic information does not help
at all. Epilepsia 2008;49(Suppl. 9):13–24.
61. Medina MT, Martinez-Juarez IE, Duron RM, et al. Treatment of myoclonic epilepsies of childhood, adolescence, and adulthood. Adv Neurol
2005;95:307–323.
62. Boillot M, Baulac S. Genetic models of focal epilepsies. J Neurosci
Methods 2015 Jun 11. pii: S0165-0270(15)00220-4. doi: 10.1016/j.
jneumeth.2015.06.003.
63. Salzmann A, Malafosse A. Genetics of temporal lobe epilepsy: a
review. Epilepsy Res Treat 2012;2012:863702.
64. Marini C, Guerrini R. The role of the nicotinic acetylcholine receptors
in sleep-related epilepsy. Biochem Pharmacol 2007;74:1308–1314.
65. Ishida S, Picard F, Rudolf G, et al. Mutations of DEPDC5 cause autosomal dominant focal epilepsies. Nat Genet 2013;45:552–555.
66. Møller RS, Heron SE, Larsen LH, et al. Mutations in KCNT1 cause a
spectrum of focal epilepsies. Epilepsia 2015 56(9):e114–e120.
67. Boutry-Kryza N, Labalme A, Ville D, et al. Molecular characterization
of a cohort of 73 patients with infantile spasms syndrome. Eur J Med
Genet 2015;58:51–58.
68. Veeramah KR, Johnstone L, Karafet TM, et al. Exome sequencing
reveals new causal mutations in children with epileptic encephalopathies. Epilepsia 2013;54:1270–1281.
69. Wilmshurst JM, Gaillard WD, Vinayan KP, et al. Summary of recommendations for the management of infantile seizures: Task Force Report
for the ILAE Commission of Pediatrics. Epilepsia 2015;56:1185–1197.
70. Nissenkorn A, Michelson M, Ben-Zeev B, et al. Inborn errors of metabolism: a cause of abnormal brain development. Neurology
2001;56:1265–1272.
71. Ficicioglu C, Bearden D. Isolated neonatal seizures: when to suspect
inborn errors of metabolism. Pediatr Neurol 2011;45:283–291.
72. Paciorkowski AR, Fang M. Chromosomal microarray interpretation:
what is a child neurologist to do? Pediatr Neurol 2009;41:391–398.
73. Ezugha H, Anderson CE, Marks HG, et al. Microarray analysis in children with developmental disorder or epilepsy. Pediatr Neurol
2010;43:391–394.
74. Ottman R, Hirose S, Jain S, et al. Genetic testing in the epilepsies–report of the ILAE Genetics Commission. Epilepsia 2010;51:655–670.
75. Lemke JR, Riesch E, Scheurenbrand T, et al. Targeted next generation
sequencing as a diagnostic tool in epileptic disorders. Epilepsia
2012;53:1387–1398.
76. Ng SB, Buckingham KJ, Lee C, et al. Exome sequencing identifies the
cause of a mendelian disorder. Nat Genet 2010;42:30–35.
77. Directors ABo. Points to consider in the clinical application of genomic
sequencing. Genet Med 2012;14:759–761.
78. Soden SE, Saunders CJ, Willig LK, et al. Effectiveness of exome and
genome sequencing guided by acuity of illness for diagnosis of neurodevelopmental disorders. Sci Transl Med 2014;6:265ra168.
79. Dixon-Salazar TJ, Silhavy JL, Udpa N, et al. Exome sequencing can
improve diagnosis and alter patient management. Sci Transl Med
2012;4:138ra178.
80. Flanagan SE, Patch AM, Ellard S. Using SIFT and PolyPhen to predict
loss-of-function and gain-of-function mutations. Genet Test Mol
Biomarkers 2010;14:533–537.