Original Article

Journal of Child Neurology

2020, Vol. 35(7) 433-441
Leukodystrophies and Genetic ª The Author(s) 2020
Article reuse guidelines:
sagepub.com/journals-permissions
Leukoencephalopathies in Children DOI: 10.1177/0883073820904294
journals.sagepub.com/home/jcn
Specified by Exome Sequencing in an
Expanded Gene Panel

Bindu Parayil Sankaran, FRACP, PhD1,2 , Madhu Nagappa, MD, DM1,2,

Shwetha Chiplunkar, PhD2,3, Sonam Kothari, PhD2,3,
Periyasamy Govindaraj, PhD2, Sanjib Sinha, MD, DM1,
and Arun B. Taly, MD, DM1,2

Abstract
The overlapping clinical and neuroimaging phenotypes of leukodystrophies pose a diagnostic challenge to both clinicians and
researchers alike. Studies on the application of exome sequencing in the diagnosis of leukodystrophies are emerging. We used
targeted gene panel sequencing of 6440 genes to investigate the genetic etiology in a cohort of 50 children with neuroimaging
diagnosis of leukodystrophy/genetic leukoencephalopathy of unknown etiology. These 50 patients without a definite biochemical
or genetic diagnosis were derived from a cohort of 88 patients seen during a 2.5-year period (2015 January-2017 June). Patients
who had diagnosis by biochemical or biopsy confirmation (n ¼ 17) and patients with incomplete data or lack of follow-up (n ¼ 21)
were excluded. Exome sequencing identified variants in 30 (60%) patients, which included pathogenic or likely pathogenic variants
in 28 and variants of unknown significance in 2. Among the patients with pathogenic or likely pathogenic variants, classic leu-
kodystrophies constituted 13 (26%) and genetic leukoencephalopathies 15 (30%). The clinical and magnetic resonance imaging
(MRI) findings and genetic features of the identified disorders are discussed.

Keywords
leukodystrophy, leukoencephalopathies, gene panel, magnetic resonance imaging, exome sequencing

Received June 9, 2019. Received revised December 30, 2019. Accepted for publication January 7, 2020.

Leukodystrophies and genetic leukoencephalopathies are autosomal recessive susceptibilities is likely to be higher as
groups of rare inherited disorders that primarily affect the demonstrated in other studies.9 In this study, we used tar-
central nervous system white matter.1,2 The overlapping clin- geted gene panel sequencing of 6440 genes to evaluate a
ical and neuroimaging phenotypes make the traditional diag- cohort of 50 children with a neuroimaging diagnosis of leu-
nostic process challenging and time consuming.3 There are kodystrophy with a high incidence of identity by descent.
about 91 disorders that are classified as leukodystrophies and The diagnostic yield and etiologic spectrum in these disorders
leukoencephalopathies.1 In the last few decades, magnetic were reviewed.
resonance imaging (MRI) pattern recognition and the advent
of next-generation technology have substantially contributed
to the identification and description of many novel
1
leukodystrophies.4 Department of Neurology, National Institute of Mental Health and
Neurosciences, Bangalore, India
Inherited white matter disorders result in significant mor- 2
Neuromuscular Lab, National Institute of Mental Health and Neurosciences,
bidity and mortality, and the children and their families expe- Bangalore, India
rience lengthy diagnostic processes.5,6 It has been shown that 3
Department of Clinical Neurosciences, National Institute of Mental Health
clinical integration of exome sequencing may decrease the and Neurosciences, Bangalore, India
number of patients with unsolved genetic white matter dis-
Corresponding Author:
orders in both children and adults.2,7,8 The diagnostic rate of Bindu Parayil Sankaran, FRACP, PhD, Genetic Metabolic Disorders Service
leukodystrophies by exome sequencing in culturally isolated (GMDS) Children’s Hospital at Westmead, Westmead NSW 2145 Australia.
communities with high incidence of identity by descent and Email: [email protected]
434 Journal of Child Neurology 35(7)

Total No. of children with leukodystrophy seen during the study period- 88

Incomplete data (n=21)

Biochemical/Biopsy diagnosis (n=17)

MLD-5,
Krabbes leukodystrophy-3
ALD-9,
Canavan’s disease-4,
Alexander’s disease-1,
Labrune’s disease-1

Undiagnosed leukodystrophies (n=50)

Exome sequencing

Pathogenic or likely pathogenic variants-28 Variants of unknown significance-2

Classic Leukodystrophies (n=13)

Genetic Leukoencephalopathies (n=15)

Figure 1. Flowchart showing the workflow for patients with leukodystrophies.

Patients and Methods MRI pattern recognition, were enrolled and evaluated as part of
This retrospective study was carried out with institutional ethical the phenotype-genotype correlative study on respiratory chain
approval. The study included a select group of 50 patients, with the disorders. All patients also underwent nerve conduction studies,
neuroimaging diagnosis of leukodystrophy/genetic leukoencephalopa- electroencephalography, and evoked potential studies as part of
thy of unknown etiology evaluated over a period of 2.5 years [2015 the diagnostic evaluation.
January–2017 June]. The primary inclusion criteria was the presence During the study period, 88 children with a neuroimaging diag-
of T2-weighted hyperintense signal changes in cerebral, cerebel- nosis of leukodystrophy were seen by the clinical team (BPS, MN,
lar, or spinal cord white matter along with clinical features SS, and ABT). Patients who had a diagnosis by biochemical or
related to white matter involvement such as pyramidal signs, biopsy confirmation (n ¼ 17; Metachromatic leukodystrophy
ataxia, and optic atrophy. Based on the clinical and MRI pattern (MLD)–5, Krabbe’s leukodystrophy–3, Adrenoleukodystrophy
recognition, patients underwent a battery of tests that included a (ALD)-9, Canavan disease–4, Alexander’s disease–1, Labrune’s
variable combination of serum ammonia, lactate, homocysteine, diseae–1) and patients with incomplete data or lack of follow-up
uric acid, amino acids, and acylcarnitine profiles, urine organic (n ¼ 21) were excluded. Fifty patients without a definite biochem-
acids estimation, lysosomal enzyme testing, very long chain fatty ical or genetic diagnosis underwent targeted gene panel sequen-
acid levels, serum Cu, ceruloplasmin, electron microscopic stud- cing. This also included patients in whom the MRI pattern
ies of axillary skin biopsy for presence of inclusions, and sural recognition provided a diagnosis, for example, vanishing white
nerve biopsy. When a mitochondrial etiology was suspected, matter disease and Alexander disease. Exome sequencing was pre-
patients underwent muscle biopsy for histopathology, spectropho- ferred over single gene sequencing in view of the easy availability
tometry for respiratory chain assays, complete mitochondrial locally of exome sequencing compared with single gene sequen-
DNA sequencing, and long range polymerase chain reaction for cing. Written informed consent was obtained for genetic analysis
mitochondrial DNA deletions in addition. Patients with mitochon- and the study from the parents or caregivers. The entire workflow
drial leukoencephalopathy, based on clinical, laboratory, and is depicted in Figure 1.
Parayil Sankaran et al 435

Exome Sequencing subclassified according to the definitions by Vanderver

et al1,7 into 3 groups: group I—disorders categorized as pri-
Samples were analyzed in a laboratory (MedGenome Labs Ltd,
mary/classic leukodystrophies; group II—genetic leukoence-
Bangalore, India) offering clinical exome sequencing as a diag-
phalopathies, which included systemic diseases that affect the
nostic service. The exome sequencing panel used in this study
white matter of the brain; and Group III, which included var-
consisted of 6440 genes. The clinical exome is a custom exome
iants of unknown significance identified in patients with
panel designed by the company (MedGenome Lab Ltd). It has
leukodystrophy.
the most relevant disease-causing genes curated from OMIM as
well as other databases like HGMD, ClinVar, and SwissVar.
The genes in which disease-causing or pathogenic variants
have been reported have been included in the panel. DNA Group 1
extracted from blood of the subjects was used to perform tar- This consisted of 13 patients (26%) with variants in known
geted gene capture using a custom capture kit. The libraries leukodystrophy genes (Table 1). The genetic findings included
were sequenced to mean >80-100X coverage on Illumina variations in EIF2B2, 3,4; GFAP; RNASEH2C; PSAP;
sequencing platform. The sequences obtained were aligned to POLR3B; CLCN2; ALDH3A2 and MLC1. The clinical and MRI
the human reference genome (GRCh37/hg19) using BWA pro- features of all patients were consistent with those reported in
gram10,11 and analyzed using Picard and GATK-Lite toolkit to literature. The patient with ALDH3A2 variation has been
identify variants relevant to the clinical indication.12,13 The reported earlier.16 Complete details of the clinical, MRI and
GATK best practices framework was followed for identifi- genetic findings of these patients are described in the Supple-
cation of variants in the sample. Gene annotation of the mentary Material, “Case Reports.”
variants was performed using VEP program against the
Ensemble release 75 human gene model.14 Clinically rele- EIF2B-related disorder. The most frequent diagnosis in group 1
vant mutations were annotated using published variants in was vanishing white matter disease. Five patients with muta-
literature and a set of diseases databases—ClinVar, OMIM, tions in EIF2B2, 3, and 4 were identified. All had episodic
GWAS, HGMD and SwissVar. Common variants were fil- neurologic regression, encephalopathy, and seizures. Their
tered based on allele frequency in 1000Genome Phase 3, MRI showed abnormal T2 signals involving all the zones of
ExAC, EVS, dbSNP141, 1000 Japanese Genome and the white matter with evidence of progressive rarefaction and cys-
internal Indian population database in the testing lab. Non- tic degeneration on fluid-attenuated inversion recovery.
synonymous variants’ effect was calculated using multiple
algorithms such as PolyPhen-2, SIFT, LRT, Mutation
CLCN2 [Chloride Ion Channel 2 (ClC-2)]-related leukoencephalopathy
Taster, and Mutation Assessor. Only nonsynonymous and
with intramyelinic edema. Clinical features in this child with
splice site variants found in the clinical exome panel were
CLCN2 variation were characterized by episodic headache,
used for clinical interpretation. Silent variations that do not
sensorineural hearing loss, and vertigo. MRI demonstrated
result in any change in amino acid in the coding region were
T2/fluid-attenuated inversion recovery hyperintense signal
not reported. Genetic test results were reported based on
changes in posterior limb of internal capsule, brain stem, and
the recommendations of American College of Medical
cerebellum (Figure 2).
Genetics.15
Exome sequencing was performed on the proband. All var-
Metachromatic leukodystrophy due to SAPOSIN B deficiency. This
iants identified by exome sequencing were verified by Sanger
4-year-old boy presented with progressive neurologic dete-
sequencing and further validated by segregation analysis in
rioration starting from 9 months of age. Evaluation showed
first-degree relatives except in 6 (Adopted patients-2; Lost to
severe demyelinating neuropathy simulating Metachromatic
follow-up-4). In patients with a family history only index
leukodystrophy, but normal Arylsulfatase A levels. MRI
patient was included for final analysis.
revealed T2/fluid-attenuated inversion recovery signal changes
in the periventricular and deep white matter with contrast
enhancement of the cranial nerves. Urinary sulfatides were not
Results measured.
During the study period, a total of 50 patients (age at evalua-
tion: range, 11 months–16 years; male-female, 1.2:1) with a Aicardi-Goutières syndrome due to RNASEH2C. This girl child
diagnosis of leukodystrophy/genetic leukoencephalopathy presented with history of developmental delay, excessive irrit-
underwent targeted gene panel sequencing. Identity by descent ability, and flexor spasms from 3 months of age. Evaluation at
was present in 21 patients (first cousin marriage, 13; maternal the age of 5 years showed severe psychomotor delay, spastic
uncle-niece marriage, 8) and family history of similar pheno- quadriparesis, and persisting tonic seizures. MRI showed cere-
types was noted in 15 (33%). bral atrophy, bilateral anterior temporal lobe cysts, symmetri-
Exome sequencing identified variants in 30 (60%) patients cal T2/fluid-attenuated inversion recovery signal changes in
which included pathogenic or likely pathogenic variants in 28 supratentorial white matter, ventricular dilation, and basal
and variants of unknown significance in 2. The cohort was ganglia calcification (Figure 3).
436 Journal of Child Neurology 35(7)

Table 1. Pathogenic and Likely Pathogenic Variants Identified in Classic Leukodystrophy Genes (Group I, n ¼ 13).
Age/
Pt no. gender Consanguinity Gene Zygosity Transcript CDS Protein Diagnosis

LE-02 7 y/M Yes EIF2B2 Hom. ENST00000266126 c.889T>A p.F297I LE with vanishing white matter
(OMIM603896)
LE-05 12 mo/F Yes EIF2B4 Hom. ENST00000451130 c.1459C>T p.R487W LE with vanishing white matter
(OMIM603896)
LE-09 8 y/M Nil EIF2B2 Hom. ENST00000266126 c.1A>G p.M1V LE with vanishing white matter
(OMIM603896)
LE-14 4 y/F Nil RNASEH2C Hom. ENST00000308418 c.205C>T p.R69W Aicardi-Goutières syndrome-3
(OMIM610329)
LE-19 4 y/M Nil ALDH3A2 Com.Het ENST00000339618 c.529C>C/Tþ p.R177Ter þ Sjogren-Larsson syndrome
c.126delG p.T43RfsTer64 (OMIM270200)

LE-21 10 y/F Yes EIF2B3 Hom. ENST00000360403 c.391T>G p.L131V LE with vanishing white matter
(OMIM603896)
LE-34 4 y/M Yes PSAP Hom. ENST00000394936 c.679_681delAAG p. K227del Metachromatic leukodystrophy
due to saposin B deficiency
(OMIM249900)
LE-44 14 mo/M Nil GFAP Het. ENST00000586793 c.716G>G p.R239H Alexander disease
(OMIM203450)
LE-58 7 y/F NA POLR3B Hom. ENST00000228347 c.2303G>A p.R768H Hypomyelinating leukodystrophy-8
(OMIM614381)
LE-64 11 y/M NA CLCN2 Hom. ENST00000265593 c.1100C>T p.P367L Leukoencephalopathy with ataxia
(OMIM615651)
LE-70 2 y/F Nil EIF2B3 Com.Het ENST00000360403 c.687T>T/G þ p.I229Mþ LE with vanishing white matter
c.956A>G p.Y319C (OMIM603896)

LE-87 11 mo/M Nil GFAP Het. ENST0000058679 c.716G>G/A p.R239 H Alexander disease
(OMIM203450)
LE-88 4 y/F Nil MLC1 Com.Het ENST00000311597 c.135dupC þ p.C46LfsTer3þ Megalencephalic
c.254G>G p.C85Y leukoencephalopathy
with subcortical cysts-1
(OMIM604004)

Abbreviations: CDS, coding DNA sequence; ENST, Ensembl gene spliced transcript; Com.Het, compound heterozygous; Hom, homozygous.

Figure 2. MRI in a child with CLCN2 mutation. Axial T2-weighted (A and B) and coronal (C) images showing bilateral symmetrical hyperintense
signal changes involving posterior limb of internal capsule, brain stem, and cerebellar white matter.

Group II chaperone HSPD. One child had variations in AUH causing

There were 15 patients (30%) with variations in genes 3-methylglutaconic aciduria, type 1. All patients with mito-
associated with genetic leukoencephalopathies, with the chondrial leukoencephalopathy had MRI features consistent
majority having variations in genes implicated in mitochon- with mitochondrial leukodystrophy, which included multifocal
drial disorders (Table 2). Variants were detected in complex confluent signal changes with restricted diffusion, contrast
1 nuclear genes (NDUFV1; NDUFS1; NDUFS2; NDUFA1 and enhancement, cysts inside the affected white matter and lactate
NDUFS8); MPV17; BOLA3; IBA57; and mitochondrial peak on magnetic resonance spectroscopy (Figure 4) as defined
Parayil Sankaran et al 437

Figure 3. Magnetic resonance image (MRI) and CT in a child with Aicardi-Goutières syndrome secondary to RNASEH2C mutation. (A) Axial T2-
weighted image shows white matter signal changes, atrophy, and ventricular dilatation. (B) Axial fluid-attenuated inversion recovery images
demonstrating anterior temporal lobe cysts. (C) Susceptibility-weighted imaging shows mineralization in dentate D. Axial CT scan shows
bilateral basal ganglia calcification.

Table 2. Pathogenic and Likely Pathogenic Variants Identified in Genes Associated With Genetic Leukoencephalopathies (Group II, n ¼ 15).
Age/
Pt no. gender Consanguinity Gene Zygosity Transcript CDS Protein change Final diagnosis

LE-04 10 y/F Yes AUH Hom. ENST00000375731 c.505þ1G>C 50 Splice site 3-Methylglutaconic aciduria type I
(OMIM250950)
LE- 06 6 y/F Yes MPV17 Hom. ENST00000380044 c.148C>T p.Arg50Trp Mitochondrial DNA depletion
syndrome 6 (OMIM256810)
LE-10 1.5 y/M Nil IBA57 Com.Het ENST00000366711 c.738C>C/Gþ p.Asn246Lys þ Multiple mitochondrial dysfunctions
c.802C>C/T p.Arg268Cys syndrome-3 (OMIM615330)
LE-15 8 y/F Yes HSPD Hom. ENST00000388968 c.1381C>A p.Gly461Leu Hypomyelinating leukodystrophy-4
(OMIM612233)
LE-16 6 y/M Nil NDUFA1 Hemi. ENST00000371437 c.94G>C p.Gly32Arg Mitochondrial complex 1 deficiency
(OMIM 252010)
LE-17 6 y/M Yes NDUFV1 Hom. ENST00000322776 c.1156C>T p.Arg386Cys Mitochondrial complex 1 deficiency
(OMIM252010)
LE-20 7 y/M Yes LYRM7 Hom. ENST00000379380 c.25C>T p.Gln9Ter Mitochondrial complex III deficiency
nuclear type 8 (OMIM615838)
LE-22 3 y/F Yes NDUFV1 Hom. ENST00000322776 c.1156C>T p.Arg386Cys Mitochondrial complex 1 deficiency
(OMIM252010)
LE-24 13 mo/M Nil NDUFS1 Com.Het ENST00000367993 c.236A>A/Cþ p.Asn79Thr þ Mitochondrial complex 1 deficiency
c.1324C>C/T p.His442Tyr (OMIM252010)
LE -25 2 y/F Yes BOLA3 Hom. ENST00000327428 c.295C>T p.Arg99Trp Multiple mitochondrial dysfunctions
syndrome-2 with
hyperglycinemia
(OMIM614299)
LE-35 2 y/F Nil NDUFV1 Hom. ENST00000322776 c.1156C>Tþ p.Arg386Cys þ Mitochondrial complex 1 deficiency
þNDUFS8 Hom. ENST00000313468 c.598G>A p.Ala200Thr (OMIM252010)
LE-48 2 y/F Yes LYRM7 Hom. ENST00000379380 c.2T>C p.Met1Thr Mitochondrial complex III deficiency
nuclear type 8 (OMIM615838)
LE-72 2 y/M Yes NDUFV1 Hom. ENST00000322776 c.1156C>T p.Arg386Cys Mitochondrial complex 1 deficiency
(OMIM252010)
LE-74 3 y/F Yes NDUFS1 Hom. ENST00000455934 c.2032T>G p.Tyr678Asp Mitochondrial complex 1 deficiency
(OMIM252010)
LE-83 2 y/M Yes IBA57 Com.Het ENST00000366711 c.706C>C/Tþ p.Pro236ser þ Multiple mitochondrial dysfunctions
c.656A>A p.Tyr219Cys syndrome-3 (OMIM615330)

Abbreviations: CDS, coding DNA sequence; ENST, Ensembl gene spliced transcript; Com.Het, compound heterozygous; Hom, homozygous.

by Kevelam et al. 2 In addition, the laboratory features Group III

and respiratory chain enzyme assays results supported diag- It consisted of variants of unknown significance identified in
noses of definite mitochondrial disorders. The MRI features patients with leukoencephalopathies. These included variation
and the steroid responsiveness in this group have been in SZT2 (p. R2435 W, Homozygous) and WDR81 (p. V107 M,
described earlier.17 Homozygous).
438 Journal of Child Neurology 35(7)

Figure 4. MRI in a child with mitochondrial leukoencephalopathy due to variation in NDUFV1 gene. (A and C) Axial and (B) coronal T2-weighted
images showing rarefied supratentorial white matter and involvement of basal ganglia and substantia nigra. (D) Fluid-attenuated inversion
recovery axial image shows cysts inside the affected white matter. (E) Diffusion-weighted image shows diffusion restriction. (F) Contrast
enhancement of the genu of corpus callosum (arrow). (G) Magnetic resonance spectroscopy shows lactate peak (arrow).

SZT2 (Seizure Threshold 2) Related Disorder constituted 13 (46%) and genetic leukoencephalopathies 15
(53.5%). Compared with similar studies in children as well as
Child with variation in SZT2 presented at the age of 10 months
in adults, the diagnostic rate is high in this study. In the study
with 1 episode of seizure. He had normal developmental mile-
by Vanderver et al,7 exome sequencing yielded a diagnostic
stones. Neurologic examination was normal except for brisk
rate of 42%. In a study on adults with leukoencephalopathy, the
stretch reflexes in the lower limbs. MRI demonstrated focal and
diagnostic yield was 26%.8 The possible reason for a high
confluent T2/fluid-attenuated inversion recovery white matter
diagnostic yield in the present study is the high degree of iden-
signal changes predominant in the posterior periventricular
tity by descent in the region.18 The high rate of identity by
region. The detected variation segregated within the family.
descent of 42% in the present study is comparable to our pre-
vious studies on metachromatic leukodystrophy and mitochon-
WDR81 drial oxidative phosphorylation disorders, which recorded
The child with WDR81 variation was born to consanguineous consanguineous rates of 62.5% and 45%, respectively.19,20 The
parents and presented with a history of impaired vision from high yield of exome sequencing in neurogenetic disorders from
the age of 5 years and gait ataxia since 13 years of age. Evalua- consanguineous communities has been reported earlier.9 Many
tion revealed optic atrophy, ataxia, and pyramidal signs. His leukodystrophies are autosomal recessive disorders, and a high
elder sister had the same phenotype, a progressive course, and degree of identity by descent might have facilitated the variant
expired at 13 years of age. MRI in both of them revealed signal calling. A high success rate has also been reported in studies
changes in the periventricular white matter, brain stem, and using research-grade exome sequencing in multiplex consan-
spinal cord. Both had nephrocalcinosis in addition. guineous families with genetic disorders.21
The etiologic spectrum in the known leukodystrophies was
similar to other studies, vanishing white matter disease being
Discussion the most common. In the study by Vanderver et al,7 the varia-
In a cohort of 50 children with neuroimaging diagnosis of tions in known leukodystrophy genes formed only a small per-
leukodystrophy/genetic leukoencephalopathy, exome sequen- centage and the majority had genetic leukoencephalopathies.
cing identified genetic variants in 60% of patients. This The primary evaluation in the study by Vanderver et al also
included pathogenic or likely pathogenic variants in 28 and included sequencing for EIF2B and GFAP genes, facilitated by
variants of unknown significance in 2. In the category of patho- MRI pattern recognition, whereas our study did not include
genic or likely pathogenic variants, classic leukodystrophies individual sequencing of these genes. The mutational spectrum
Parayil Sankaran et al 439

of vanishing white matter disease being wide, next-generation significance of the combination of WDR81 variation in the
sequencing has revealed variation in 3 different genes causing patient with leukoencephalopathy and nephrocalcinosis is
the disease in 5 patients in the present study.22 unclear at present. Homozygous mutations in WDR81 have
The identification of rare disorders such as CLCN2-related been associated with cerebellar ataxia, mental retardation, and
disorder and saposin deficiency exemplifies the role of exome disequilibrium syndrome (CAMRQ).28 These 2 variants of
sequencing in the diagnosis of rare leukodystrophies. CLCN2- unknown significance require functional analysis to prove their
related disorder is characterized by nonspecific neurologic possible pathogenic role.
findings and is reported to be rare in children.23 The clinical The impact of exome sequencing has been determined in
features in children include mild learning disability, headache, terms of the expenses incurred and time to diagnosis in previ-
action tremor, and gait ataxia.24 Episodic headache and sensor- ous studies.6,9 Richards et al6 has concluded that appropriately
ineural hearing loss and vertigo as reported in our patient have incorporating next-generation sequencing (NGS) into diagnos-
also been described. The characteristic MRI findings include tic algorithms may result in lower charges, reduced time to
prominent signal abnormalities and decreased apparent diffu- diagnosis, and reduced amount of testing, including reducing
sion coefficient values in the posterior limb of the internal repeat MRIs and associated sedations. However, a significant
capsule, cerebral peduncle, and pyramidal tracts in pons. Addi- number of children with white matter disorders still remained
tional features include cerebellar white matter hyperintensities without a diagnosis, highlighting the complexity and the
and features simulating hypomyelination in cerebral white mat- unknown genes involved in white matter disease pathology.
ter as observed in our patient.24 It remains to be seen if these children could be diagnosed by
The high incidence of mitochondrial leukoencephalopathy use of whole exome sequencing or whole genome sequencing
in the group of genetic leukoencephalopathies is noteworthy. rather than targeted exome sequencing. The diagnostic super-
The role of NGS in facilitating molecular diagnosis in disor- iority of whole genome sequencing over whole exome sequen-
ders with heterogeneous genetic etiology such as hypomyeli- cing includes its ability to detect intronic single-nucleotide
nation and respiratory chain disorders has been emphasized variants, single-nucleotide variants in noncoding RNA, small
earlier.2 It is important to recognize that the presentation in copy number variations, mitochondrial mutations, as well as
these disorders were primarily with features of leukodystro- exonic single-nucleotide variants missed because of incom-
phy rather than with systemic features. In addition to complex plete coverage and enrichment methods.29 Whole genome
1 nuclear genes, the study also identified variations in genes sequencing is being established/recommended as a first tier test
causing combined phosphorylation disorder and MitCHAP-60 in clinically heterogeneous cohorts, despite its higher cost and
disease. HSPD1, a mitochondrial Hsp60 chaperon in encoding increased requirements for data analysis and storage.29,30 Nev-
the conserved heat shock protein 60, has been previously ertheless, the cost of NGS is significantly decreasing, and it is
identified in a large inbred Israeli Bedouin kindred with a envisaged that whole genome sequencing will become the stan-
lethal hypomyelinating leukodystrophy. 25 Identification of dard diagnostic tool to detect both copy number variations and
the HSPD1 mutations in a consanguineous family in our study point mutations in the near future. Studies have also proven that
further confirms this association. Also noteworthy is the sta- periodic reanalysis of the data in individuals not diagnosed on
ble clinical course in this group. After the initial 1 or 2 epi- initial testing could improve the diagnostic yield.31,32 How-
sodes of regression in early life, the majority of the patients ever, in the study by Lynch et al,8 resequencing the negative
remained stable with spastic paraparesis or quadriparesis. patients with whole exome did not improve the diagnostic
This implies that the disability and cost of medical care in yield. It is also emphasized that a negative result should always
genetic encephalopathies may be similar to classic prompt the physician to pursue alternate diagnostic approaches
leukodystrophies. if the phenotype is convincing.33
Leukoencephalopathy was also demonstrated in a child with In conclusion this study identified a high diagnostic rate in
3-methylglutaconic aciduria type 1, with seizures, learning dis- patients with a clinically and radiologically ascertained cohort
ability, and pyramidal signs. MRI features included focal dis- of children with a neuroimaging diagnosis of leukodystrophy/
crete subcortical white matter hyperintensities. AH1 mutations genetic leukoencephalopathies from a population with a high
have been previously linked to adult-onset leukoencephalopa- incidence of identity by descent. Genetic leukoencephalopa-
thy.26,27 Mild abnormalities in deep supratentorial white matter thies emerged as an important cause of white matter disorders
have been described in a 10-year-old child.26 The authors pro- in children emphasizing that the burden of genetic leukoence-
posed that these abnormalities represent the earliest stages of phalopathies may be similar to that of the classic leukodystro-
the slowly progressive neurodegenerative disorder, mainly phies. Exome sequencing also can lead to identification of new
affecting the white matter, observed in adult patients. It would genes associated with leukoencephalopathy, paving the way for
be worthwhile to add this disorder in the group of genetic future studies and better diagnosis.
leukoencephalopathies.1
Variants of uncertain significance were identified in 2 Author’s Note
genes, SZT2 and WDR81. The SZT2 gene variation was also Periyasamy Govindaraj is also affiliated with Institute of Bioinfor-
identified in the cohort described by Vanderver et al,7 suggest- matics, International Tech Park, Bangalore, India and Manipal Acad-
ing its possible association with leukoencephalopathy. The emy of Higher Education, Manipal, India.
440 Journal of Child Neurology 35(7)

Acknowledgment 7. Vanderver A, Simons C, Helman G, et al. Whole exome sequen-

We acknowledge MedGenome Labs Ltd, Bangalore, India, for carry- cing in patients with white matter abnormalities. Ann Neurol.
ing out the exome sequencing for the patients. 2016;79(6):1031-1037.
8. Lynch DS, Rodrigues Brandao de Paiva A, Zhang WJ, et al. Clin-
Author Contributions ical and genetic characterization of leukoencephalopathies in
adults. Brain. 2017;40(5):1204-1211.
BPS and ABT were responsible for the study concept and design and
9. Yavarna T, Al-Dewik N, Al-Mureikhi M, et al. High diagnostic
acquisition, analysis, and interpretation of data. BPS performed the
yield of clinical exome sequencing in Middle Eastern patients
literature search and wrote the manuscript. ABT, BPS, MN, SS, SC,
and KS comprised the clinical team that was involved in the evalua- with Mendelian disorders. Hum Genet. 2015;134(9):967-980.
tion, management, and follow-up of the patients. KS acquired and 10. Li H, Durbin R. Fast and accurate long-read alignment with
interpreted the respiratory chain assays. KS, SC, and PG performed Burrows-Wheeler transform. Bioinformatics. 2010;26(5):
the mitochondrial DNA sequencing and long-range polymerase chain 589-595.
reaction for mitochondrial DNA deletions. All authors reviewed and 11. Meyer LR, Zweig AS, Hinrichs AS, et al. The UCSC Genome
approved the final manuscript Browser database: extensions and updates 2013. Nucleic Acids
Res. 2013;41(Database issue):D64-D69.
Declaration of Conflicting Interests 12. McKenna A, Hanna M, Banks E, et al. The Genome Analysis
The authors declared no potential conflicts of interest with respect to Toolkit: a MapReduce framework for analyzing next-generation
the research, authorship, and/or publication of this article. DNA sequencing data. Genome Res. 2010;20(9):1297-1303.
13. Li H, Handsaker B, Wysoker A, et al. The sequence alignment/
Funding map format and SAMtools. Bioinformatics. 2009;25(16):
The authors disclosed receipt of the following financial support for the 2078-2079.
research, authorship, and/or publication of this article: This study was 14. McLaren W, Pritchard B, Rios D, Chen Y, Flicek P, Cunningham
supported in part by a grant from Indian council of Medical Research F. Deriving the consequences of genomic variants with the
to BPS (Grant No. 54/9/2012-HUM-BMS). Ensembl API and SNP Effect Predictor. Bioinformatics. 2010;
26(16):2069-2070.
ORCID iD 15. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the
Bindu Parayil Sankaran, FRACP, PhD https://orcid.org/0000- interpretation of sequence variants: a joint consensus recommen-
0002-1914-9505 dation of the American College of Medical Genetics and Geno-
mics and the Association for Molecular Pathology. Genet Med.
Supplemental Material 2015;17(5):405-424.
16. Nagappa M, Bindu PS, Chiplunkar S, et al. Sjögren-Larsson syn-
Supplemental material for this article is available online.
drome. Neurology. 2017;88(1):e1-e4.
17. Bindu PS, Sonam K, Chiplunkar S, et al. Mitochondrial leukoen-
Ethical Approval
cephalopathies: A border zone between acquired and inherited
This study was approved by the Institutional Ethics Committee [no. white matter disorders in children? Mult Scler Relat Disord.
NIMHANS/105th IEC/2016]. 2018;20:84-92.
18. Bittles AH, Devi AR, Rao NA. Consanguinity, twinning and sec-
References ondary sex ratio in the population of Karnataka, south India. Ann
1. Vanderver A, Prust M, Tonduti D, et al. Case definition and Hum Biol. 1988;15(6):455-460.
classification of leukodystrophies and leukoencephalopathies. 19. Bindu PS, Mahadevan A, Taly AB, Christopher R, Gayathri N,
Mol Genet Metab. 2015;114(4):494-500. Shankar SK. Peripheral neuropathy in metachromatic leucody-
2. Kevelam SH, Steenweg ME, Srivastava S, et al. Update on leu- strophy. A study of 40 cases from south India. J Neurol Neurosurg
kodystrophies: a historical perspective and adapted definition. Psychiatry. 2005;76(12):1698-1701.
Neuropediatrics. 2016;47(6):349-354. 20. Sonam K, Bindu PS, Srinivas Bharath MM, et al. Mitochondrial
3. Parikh S, Bernard G, Leventer RJ, et al. A clinical approach to the oxidative phosphorylation disorders in children: phenotypic, gen-
diagnosis of patients with leukodystrophies and genetic leukoen- otypic and biochemical correlations in 85 patients from South
cephalopathies. Mol Genet Metab. 2015;114(4):501-515. India. Mitochondrion. 2017;32:42-49.
4. Ashrafi MR, Tavasoli AR. Childhood leukodystrophies: a litera- 21. Alazami AM, Patel N, Shamseldin HE, et al. Accelerating novel
ture review of updates on new definitions, classification, diagnos- candidate gene discovery in neurogenetic disorders via whole-
tic approach and management. Brain Dev. 2017;39(5):369-385. exome sequencing of prescreened multiplex consanguineous fam-
5. Bonkowsky JL, Nelson C, Kingston JL, Filloux FM, Mundorff ilies. Cell Rep. 2015;10(2):148-161.
MB, Srivastava R. The burden of inherited leukodystrophies in 22. Pronk JC, van Kollenburg B, Scheper GC, van der Knaap MS.
children. Neurology. 2010;75(8):718-725. Vanishing white matter disease: a review with focus on its genet-
6. Richards J, Korgenski EK, Taft RJ, Vanderver A, Bonkowsky JL. ics. Ment Retard Dev Disabil Rres Rev. 2006;12(2):123-128.
Targeted leukodystrophy diagnosis based on charges and yields 23. van der Knaap MS, Depienne C, Sedel F, Abbink TEM. CLCN2-
for testing. Am J Med Genet A. 2015;167A(11):2541-2543. related leukoencephalopathy. In: Pagon RA, Adam MP, Ardinger
Parayil Sankaran et al 441

HH, et al, eds. GeneReviews(R). Seattle, WA: University of 29. Lionel AC, Costain G, Monfared N, et al. Improved diagnostic
Washington; 1993. yield compared with targeted gene sequencing panels suggests a
24. Depienne C, Bugiani M, Dupuits C, et al. Brain white matter role for whole-genome sequencing as a first-tier genetic test.
oedema due to ClC-2 chloride channel deficiency: an observa- Genet Med. 2018;20(4):435-443.
tional analytical study. Lancet Neurol. 2013;12(7):659-668. 30. Stavropoulos DJ, Merico D, Jobling R, et al. Whole genome
25. Magen D, Georgopoulos C, Bross P, et al. Mitochondrial hsp60 sequencing expands diagnostic utility and improves
chaperonopathy causes an autosomal-recessive neurodegenera- clinical management in pediatric medicine. NPJ Genom
tive disorder linked to brain hypomyelination and leukodystro- Med. 2016;1.
phy. Am J Hum Genet. 2008;83(1):30-42. 31. Costain G, Jobling R, Walker S, et al. Periodic reanalysis of
26. Wortmann SB, Kremer BH, Graham A, et al. 3-Methylglutaconic whole-genome sequencing data enhances the diagnostic advan-
aciduria type I redefined: a syndrome with late-onset leukoence- tage over standard clinical genetic testing. Eur J Hum Genet.
phalopathy. Neurology. 2010;75:1079-1083. 2018;26(5):740-744.
27. Eriguchi M, Mizuta H, Kurohara K, et al. 3-Methylglutaconic 32. Machini K, Ceyhan-Birsoy O, Azzariti DR, et al. Analyzing and
aciduria type I causes leukoencephalopathy of adult onset. Neu- reanalyzing the genome: findings from the MedSeq Project. Am J
rology. 2006;67(10):1895-1896. Hum Genet. 2019;105(1):177-188.
28. Gulsuner S, Tekinay AB, Doerschner K, et al. Homozygosity map- 33. Pena LDM, Jiang YH, Schoch K, et al. Looking beyond the
ping and targeted genomic sequencing reveal the gene responsible exome: a phenotype-first approach to molecular diagnostic reso-
for cerebellar hypoplasia and quadrupedal locomotion in a consan- lution in rare and undiagnosed diseases. Genet Med. 2018;20(4):
guineous kindred. Genome Res. 2011;21(12):1995-2003. 464-469.