Received: 19 August 2020 | Revised: 19 April 2021 | Accepted: 28 April 2021

DOI: 10.1002/humu.24216

RESEARCH ARTICLE

Exome sequencing as the first‐tier test for pediatric

respiratory diseases: A single‐center study

Chanjuan Hao1,2 | Ruolan Guo1,2 | Jun Liu3 | Xuyun Hu1,2 | Jun Guo1,2 |
3 3 1,2 3 3
Yao Yao | Zhipeng Zhao | Zhan Qi | Jun Yin | Lanqin Chen |
Hao Wang3 | Baoping Xu3 | Wei Li1,2

1
Beijing Key Laboratory for Genetics of Birth Defects, Beijing Pediatric Research Institute, MOE Key Laboratory of Major Diseases in Children, Genetics and Birth
Defects Control Center, Beijing Children's Hospital, National Center for Children's Health, Capital Medical University, Beijing, China
2
Henan Key Laboratory of Pediatric Inherited & Metabolic Diseases, Henan Children's Hospital, Zhengzhou Hospital of Beijing Children's Hospital,
Zhengzhou, China
3
Respiratory Department of Beijing Children's Hospital, China National Clinical Research Center of Respiratory Diseases, National Center for Children's Health,
Capital Medical University, Beijing, China

Correspondence
Chanjuan Hao and Wei Li, Beijing Key Abstract
Laboratory for Genetics of Birth Defects,
The high clinical and genetic heterogeneity makes it difficult to reach a confirmative
Beijing Pediatric Research Institute, MOE Key
Laboratory of Major Diseases in Children, diagnosis of suspected pediatric respiratory inherited diseases. Many patients with
Genetics and Birth Defects Control Center,
monogenic respiratory disorders could be missed without genetic testing. We
Beijing Children's Hospital, National Center for
Children's Health, Capital Medical University, performed a single‐center study in Beijing Children's Hospital to demonstrate the
100045 Beijing, China.
clinical utility of exome sequencing (ES) as a first‐tier test by evaluating the diag-
Email: hchjhchj@163.com and
liwei@bch.com.cn nostic yields of ES for inherited diseases with respiratory symptoms. A total of
Baoping Xu, China National Clinical Research 107 patients were recruited in this study. We identified 51 pathogenic or likely
Center of Respiratory Diseases, Respiratory,
pathogenic variants in 37 patients by ES (with or without copy number variants
Department of Beijing Children's Hospital,
Capital Medical University, National Center for sequencing). The overall diagnostic yield was 34.6% (37/107). The most frequent
Children's Health, 100045 Beijing, China. disorders in our cohort were primary immunodeficiency disease (PIDs) (18/37,
Email: xubaopingbch@163.com
48.6%) and primary ciliary dyskinesia (PCD) (9/37, 24.3%). We further reviewed the
Funding information directive outcomes of genetic testing on the 37 positive cases. Our study demon-
Ministry of Science and Technology of China, strated the effectiveness of ES as a first‐tier test in China for diagnosing monogenic
Grant/Award Number: 2016YFC1000306;
Special Fund of the Pediatric Medical diseases of the respiratory system. In the era of precision medicine, ES as a first‐tier
Coordinated Development Center of Beijing test can rapidly make a molecular diagnosis and direct the intervention of the
Hospitals Authority, Grant/Award Number:
XTCX201807; Beijing Municipal Science and positive cases in pediatric respiratory medicine.
Technology Commission Foundation,
Grant/Award Number: Z181100001918003; KEYWORDS
Beijing Municipal Commission of Health and
exome sequencing, first‐tier test, monogenic diseases, pediatric respiratory medicine, precision
Family Planning Foundation,
medicine
Grant/Award Numbers: 2018‐2‐1141, 2020‐4‐
1144; Beihang University & Capital Medical
University Advanced Innovation Center for Big
Data‐Based Precision Medicine Plan,
Grant/Award Number: BHME‐201905

Chanjuan Hao, Ruolan Guo, and Jun Liu contributed equally to this study. Chanjuan Hao, Ruolan Guo, and Jun Liu should be considered joint first authors.

Human Mutation. 2021;1–10. wileyonlinelibrary.com/journal/humu © 2021 Wiley Periodicals LLC | 1

2 | HAO ET AL.

1 | INTRODUCTION 2.2 | Patients

Next‐generation sequencing (NGS) is now being widely incorporated Patients in this study were initially referred to the Department of
into the genetic diagnosis of pediatric monogenic disorders Respiratory Medicine of Beijing Children's Hospital from May 2018
(Hu et al., 2018; Meng et al., 2017; Scocchia et al., 2019; Yang to January 2020. The inclusion criteria were hospitalized patients,
et al., 2013). Compared with other medical disciplines, the applica- respiratory involvement, and suspected genetic disorders. The as-
tion of genetic testing in respiratory medicine remains in its infancy. sessment for the option of genomic testing assays was carried out by
Nevertheless, advances in sequencing techniques and the knowledge a panel of trained physicians and geneticists at Beijing Children's
of genetics and genomics have already promoted the applications of Hospital. ES was performed for suspected monogenic disorders with
NGS in monogenic respiratory disorders, such as primary im- respiratory system involvement. Parallel testing by ES and copy
munodeficiency disease (PIDs) and primary ciliary dyskinesia (PCD) number variants sequencing (CNV‐seq) were performed for sus-
(Arts et al., 2019; Cifaldi et al., 2019; Fassad et al., 2020; Marshall pected syndromic disorders, as the likelihood of a chromosomal
et al., 2015; Rudilla et al., 2019). Mendelian disorders in respiratory condition was not excluded by the specialist team.
medicine mainly consist of the following three categories: (i) pul-
monary disorders, including airway disease, pulmonary parenchymal
disease, and pulmonary vascular disease; (ii) sleep disorder; (iii) other 2.3 | Exome sequencing and variant interpretation
monogenic diseases with respiratory system involvement, including pipeline
PIDs, neuromuscular diseases, and inherited metabolic disease (Yao
& Shen, 2017). The clinical manifestation of a majority of monogenic Genomic DNA was isolated from peripheral blood obtained from
respiratory disorders is often nonspecific and overlaps with other the probands and their parents using a Gentra Puregene Blood
common respiratory diseases, including cough, asthma, pneumonia, Kit (QIAGEN). Libraries of genomic DNA were captured using the
bronchitis, bronchopneumonia, and bronchiectasis. These atypical Agilent Sureselect Human All Exon v6 kit (Agilent Technologies
phenotypes pose a challenge to early diagnosis and disease man- Inc.) and were sequenced on an Illumina NovaSeq. 6000 Analyzer
agement. In this sense, obtaining a definite diagnosis in suspected (Illumina) according to the manufacturer's recommendations for
cases can be laborious and, in many occasions, ineffective (Cifaldi 150‐bp paired‐end runs. The exome sequencing resulted in over
et al., 2019; Damseh et al., 2017; Fassad et al., 2020; Marshall 12 Gb of clean data. The average sequencing depth was more
et al., 2015; Rudilla et al., 2019). than 100X. Quality score fulfilled Q20>95% and Q30>90%. Se-
Exome sequencing (ES) or genome sequencing (GS) has shown quencing reads (fastq files) were aligned to the GRCh37/hg19
advantages for expeditious detection of pediatric disorders (Sanford human reference genome sequence using Burrows‐Wheeler
et al., 2019; Scocchia et al., 2019; Stavropoulos et al., 2016). How- Aligner (BWA) and bam files were created by Picard. Genome
ever, the feasibility of NGS in pediatric respiratory diseases has not Analysis Toolkit (GATK) software was used to perform the var-
been systematically explored. In our cohort, 107 patients with sus- iant calling. Variants were annotated and filtered by TGex
pected inherited diseases were recruited and diagnostic genomic (https://geneyx.com/geneyxanalysis/). The analysis regions in-
testing was performed. Of the 107 patients tested, 34.6% (37/107) cluded all exonic nucleotides and the 50‐bp flanking intronic
received a molecular diagnosis. The two most frequent monogenic nucleotides. The main reference databases of TGex included
respiratory disorders were PIDs and PCD. This is the first single‐ population databases (dbSNP https://www.ncbi.nlm.nih.gov/snp/,
center study on the clinical application of NGS to evaluate its fea- 1000G https://www.internationalgenome.org/, gnomAD http://
sibility in pediatric monogenic respiratory diseases in China, reveal- gnomad.broadinstitute.org/) and disease databases (HGMD
ing the monogenic disease spectrum of hospitalized patients in https://portal.biobase-international.com/hgmd/, ClinVar https://
respiratory medicine. www.ncbi.nlm.nih.gov/clinvar/, OMIM https://omim.org/, and
MalaCards https://www.malacards.org/). All variants were clas-
sified according to the American College of Medical Genetics and
2 | M A TE R I A L S AN D ME TH O D S Genomics and the Association for Molecular Pathology (ACMG/
AMP) interpretation standards and guidelines (Richards
2.1 | Editorial policies and ethical considerations et al., 2015). Copy number variants (CNVs) based on exome se-
quencing data were generated using the CNV detection program:
The study was approved by the institutional review board of Beijing CNVkit. The clinical significance of CNVs was analyzed and in-
Children's Hospital (BCH, Approval No. 2015‐26). All the patients or terpreted on the basis of Database of Genomic Variants,
legal guardians provided written informed consent for this study. For DECIPHER database, ClinVar, OMIM, and ClinGen (Clinical
the patients undergoing ES tests, they or their legal guardians have Genome Resource), and subsequently classified according to
the right to decide whether or not to be informed of the results of the ACMG and ClinGen recommendation (Kearney et al., 2011;
secondary findings. Riggs et al., 2020).
HAO ET AL. | 3

2.4 | Multiplex ligation‐dependent probe (i) variants were classified by following the ACMG/AMP guidelines
amplification (MLPA) into pathogenic/likely pathogenic (P/LP) ones; (ii) variants were in
line with the segregation law (a heterozygous P/LP variant was de-
For one or more exons deletion or duplication of gene level, we tected in the gene with dominant inheritance pattern, or homo-
use multiplex ligation‐dependent probe amplification (MLPA) to zygous/compound heterozygous P/LP variants were detected in the
make a further verification of the copy number variations (CNVs). gene with recessive inheritance pattern). The overall diagnostic yield
The MLPA probes, reaction, and data analysis were performed of 107 patients was 34.6% (37/107). Among these 37 cases, two of
according to the recommendation of the MRC‐Holland protocol them (Patient 44 and Patient 73) have been previously described and
(www.mrcholland.com). published (Hu et al., 2019; Liu et al., 2019). Autosomal dominant
(AD), autosomal recessive (AR), X‐linked dominant (XLD), and
X‐linked recessive (XLR) disorders were observed in 12 (32.4%),
2.5 | Parallel testing by exome sequencing and 18 (48.6%), 1 (2.7%), and 6 (16.2%) patients, respectively (Figure 2a).
copy number variation sequencing 37.2% (19/51) SNVs and CNVs were not reported in previous lit-
erature or public databases. Of the 12 AD cases, 10 (83.3%) had de
For the patients with suspected syndromic disorders, we performed novo variants and 2 (Patient 98 and Patient 100, 16.7%) had in-
parallel ES and CNV‐seq. In this strategy, CNVs detection and calling herited variants. Reviewing the family history of Patient 98, we
were performed using our in‐house pipeline. In brief, CNVs detection found the father had recurrent epistaxis during childhood, which was
was based on two sequencing datasets. One data set was derived from a characteristic feature of ENG mutation. The father of Patient 100,
the sequencing of the targeted exome library captured by the uniquely with a mutation in SFTPC, has thus far experienced no symptoms,
designed probes. The probes (iGeneTech) were added per 50 bp in in- which is reasonable and acceptable as pulmonary surfactant meta-
tergenic regions and intronic regions to better call CNVs from ES data. bolism dysfunction‐2 (SMDP2) caused by SFTPC mutation has a
CNVkit was used to call the CNVs (>250 kb) and the parameters were highly variable phenotype and reduced penetrance. Among the
as default. The other part of the data was derived from the direct se- 18 AR disorders, there were 4 (22.2%) cases of homozygosity and
quencing of the library without capture. Each sample yielded one giga 14 (77.8%) cases of compound heterozygosity. Of the 6 XLR cases,
base (Gb) raw data. CNV analysis based on low‐coverage genome se- 1 (16.7%) variant was de novo and 5 (83.3%) were inherited from
quencing (0.3X) was used to call the large CNVs and the healthy parents maternal carriers. The XLD de novo variant was identified in
were used as the reference control samples. The procedure is described the FLNA gene in Patient 12 (female). Two CNVs were observed in
in details elsewhere (Hu et al., 2020). The analysis and interpretation of two individuals. The CNV detected in Patient 73 by parallel tests was
CNVs have been described before and were similarly carried. de novo (Hu et al., 2019). Patient 6 had a homozygous approximately
200 kb deletion (ranging chr9: 214929‐418085, GRCh37/hg19) en-
compassing exons 1–30 of the DOCK8 gene that was inherited from
3 | RESULTS the consanguineous parents. This deletion analysis was performed
based on the ES reads depth data and further confirmed by MLPA
3.1 | Demographics and test settings (using the SALSA MLPA P385‐A2 DOCK8 probe mix) (Table 1).

In the 107 patients, there were 73 males (68.2%) and 34 females

(31.8%) (sex ratio = 2.15), with a median age of 15 months (ranging from 3.3 | Precision diagnosis and its clinical
1 month to 18 years) at the time of sample collection. Detailed phe- management
notypic data were collected and anonymously shown in Tables S1
and S2 according to the standardized Human Phenotype Ontology Of the diagnosed disorders, PIDs were diagnosed in 48.6% (18/37) of
terms. Of the 107 patients, 96 (89.7%) (mainly with respiratory system the patients, which was the most frequent monogenic disorder in
involvement) were performed with ES, with 94 receiving trio‐ES and 2 respiratory medicine. The second‐most frequent disorder was PCD
(P54 and P106) proband‐ES due to unavailability of their parental accounting for 24.3% (9/37) of the patients, followed by 16.2%
samples; 11 (10.3%) with multiple congenital malformations or syn- (6/37) syndromic disorders (Figure 2b).
dromic phenotypes were performed with parallel tests of ES and The molecularly definitive diagnosis was classified into four ca-
CNV‐seq (Figure 1). The overall average turnaround time (TAT) to issue tegories (Figure 2c). A, 21.6% of patients (8/37) who had no clinical
an initial report was 4 weeks (ranging from 3 to 6 weeks). diagnosis due to nonspecific phenotypes obtained definitive diag-
nosis. Of these, five cases presented with multiple congenital mal-
formations, one case was diagnosed with interstitial lung disease
3.2 | Mutation types and inheritance (ILD), one with achondroplasia, and one with platelet glycoprotein IV
deficiency. B, 8.1% of the patients (3/37) changed the initial diagnosis
We identified 51 pathogenic or likely pathogenic variants in 37 pa- after obtaining a molecular diagnosis. C, 15 patients (40.5%) who
tients (Table 1). The criteria for establishing a positive test included: were clinically diagnosed as PIDs were molecularly confirmed and
4 | HAO ET AL.

F I G U R E 1 Flow diagram of the number of patients who were enrolled and received genetic testing by ES or parallel tests of ES and
CNV‐seq. *represented multiple congenital malformations or syndromic phenotypes. **CNVs detection and calling were performed using an in‐
house pipeline, see Section 2. CNV‐seq, copy number variants sequencing; ES, exome sequencing; Indel, small insertion and deletion; MLPA,
multiplex ligation‐dependent probe amplification; RT‐PCR, reverse transcription‐polymerase chain reaction; SNV, single‐nucleotide variation

classified into different subtypes, for which specific treatment op- For 12 patients who were molecularly confirmed to have PIDs,
tions were considered. And D, 11 patients whose previous diagnoses bone marrow transplantation was a recommended treatment. No-
were confirmed according to genetic testing results, mostly as di- tably, Patient 53 with ELANE mutation that caused severe congenital
agnoses of PCD (6/11). neutropenia (OMIM: 202700) received hematopoietic stem cell
Molecular findings guided the clinicians to reach a definitive transplantation.
diagnosis, which is helpful in patient care and deciding effective Patient 92 was a 5‐month‐old preterm‐birth boy with
therapeutic measures. A subsequent change in clinical management bronchopneumonia, abnormal facial features, hydrocephalus, and
based on the molecular diagnosis was shown in the following aspects: macrocephaly. Molecular testing identified a de novo PIK3CA
(i) treatment planning; (ii) further examination of the respiratory mutation, and Cowden syndrome 5 (OMIM: 615108) was diag-
system; (iii) referral for systemic evaluation; (iv) provision of appro- nosed (Orloff et al., 2013). Molecular findings guided the clinicians
priate genetic counseling (Table 1). Here, we list some examples. to reach a definite diagnosis, which is helpful to the proband's care
 Details of molecular findings in 37 patients who received a diagnosis
HAO

TABLE 1

Effect on
ET AL.

Variant(s) Inherited clinical Effect on clinical

[2]
ID Gene Genomic variant(s) Reference sequence classification pattern Zygosity diagnosis management [3]

Patient 06 DOCK8 Exon 1–30 deletion NM_203447.4 P AR Inherited homo C I

Patient 10 DNAH5 c.7967T>C/p.Ile2656Thr NM_001369.2 LP+LP AR Inherited D IV

c.1089del/p.Leu364Tyrfs*3[1] compound het

Patient 12 FLNA c.2527dup/p.Ala843Glyfs*12[1] NM_001456.3 P XLD De novo het A III

[1]
Patient 14 DNAAF1 c.1496del/p.Pro499Hisfs*36 NM_178452.5 P+P AR Inherited B II
c.1930A>T/p.Arg644*[1] compound het

Patient 15 DNAH5 c.9033del/p.Phe3011Leufs*17[1] NM_001369.2 P+P AR Inherited D IV

c.8806del/p.Val2936Phefs*3[1] compound het

Patient 17 DNAH11 c.7441‐1G>C[1] NM_001277115.1 P+P AR Inherited B IV

c.8769_8785del/ compound het
p.Val2924Alafs*13[1]

Patient 22 FGFR3 c.1138G>A/p.Gly380Arg NM_000142.4 P AD De novo het A III

Patient 25 CCNO c.263_267dup/p.Val90Serfs*6 NM_021147.4 P+P AR Inherited B II

c.248_252dup/p.Gly85Cysfs*11 compound het

Patient 26 STAT3 c.1699A>G/p.Asn567Asp NM_003150.3 LP AD De novo het C I

[1]
Patient 30 ELANE c.368T>C/p.Leu123Pro NM_001972.3 LP AD De novo het C I

Patient 34 RAG1 c.1677G>T/p.Arg559Ser NM_000448.2 P+P AR Inherited C I

c.2095C>T/p.Arg699Trp compound het

Patient 35 PRF1 c.769T>G/p.Cys257Gly[1] NM_005041.4 LP+LP AR Inherited C I

c.147C>A/p.Asp49Glu[1] compound het

Patient 36 ITGB2 c.1657+1G>T[1] NM_000211.4 P AR Putative homo[4] C I

Patient 44 LPIN2 c.2327+1G>C NM_014646.2 P+P AR Inherited C I

c.1691_1694del/p. Arg564Lysfs*3 compound het

Patient 50 CCDC39 c.2040_2043del/ NM_181426.1 P AR Inherited homo D II

p.Cys680Trpfs*15

Patient 53 ELANE c.640G>A/p.Gly214Arg NM_001972.3 P AD De novo het C I

Patient 56 ODAD3 c.267dup/p.Glu90* NM_145045.4 P+P AR Inherited D II

c.75del/p.Arg26Glyfs*56[1] compound het

Patient 57 DPH1 c.397T>C/p.Tyr133His NM_001383.3 LP+P AR Inherited A III

c.472del/p.Arg158Glyfs*2 compound het
|

(Continues)
5
 6

TABLE 1 (Continued)
|

Effect on
Variant(s) Inherited clinical Effect on clinical
[2]
ID Gene Genomic variant(s) Reference sequence classification pattern Zygosity diagnosis management [3]

Patient 64 BTK c.1349+1G>A NM_000061.2 P XLR Inherited hemi D IV

Patient 65 PIK3CD c.3061G>A/p.Glu1021Lys NM_005026.3 P AD De novo het C I

[1]
Patient 69 IL2RG c.695G>T/p.Gly232Val NM_000206.2 LP XLR Inherited hemi C I

Patient 73 ‐ 14q13.1‐21.1 del 3.1Mb LP AD De novo het A III

Patient 74 CYBB c.752G>A/p.Trp251* NM_000397.3 P XLR Inherited hemi D I

Patient 76 ADA c.424C>T/p.Arg142* NM_000022.3 P+P AR Inherited D I

c.467G>A/p.Arg156His compound het

Patient 80 PIK3CD c.3061G>A/p.Glu1021Lys NM_005026.3 P AD De novo het C I

Patient 82 LRBA c.4801C>T/p.Arg1601* NM_006726.4 LP+LP AR Inherited C I

c.4239del/p.Val1414Trpfs*33[1] compound het

Patient 85 FOXP3 c.1010G>A/p.Arg337Gln NM_014009.3 P XLR De novo hemi C I

[1]
Patient 87 CD36 c.1142T>G/p.Leu381* NM_000072.3 LP+LP AR Inherited A II
c.1228_1239del/ compound het
p.Ile410_Ile413del

Patient 88 DNAH11 c.3470T>G/p.Leu1157Arg NM_001277115.1 LP+P AR Inherited D II

c.9454A>T/p.Lys3152*[1] compound het

Patient 90 DNAH11 c.13373C>T/p.Pro4458Leu NM_001277115.1 P AR Inherited homo D II

Patient 92 PIK3CA c.2740G>A/p.Gly914Arg NM_006218.3 LP AD De novo het A III

Patient 93 CYBB c.190T>C/p.Cys64Arg NM_000397.3 P XLR Inherited hemi C I

Patient 98 ENG c.808C>T/p.Gln270* NM_000118.3 P AD Inherited het D IV

Patient 100 SFTPC c.218T>C/p.Ile73Thr NM_003018.3 P AD Inherited het A I

Patient 101 KMT2D c.4464del/p.Cys1489Valfs*17[1] NM_003482.3 P AD De novo het D III

Patient 103 CYBB c.1271G>A/p.Trp424* NM_000397.3 P XLR Inherited hemi C I

[1]
Patient 104 SMARCB1 c.1084_1086del/p.Glu362del NM_003073.4 LP AD De novo het A III
[1] [2]
Note: Unreported variants before. A: to produce new clinical diagnoses and guide the etiology. B: to change initial diagnoses and identify specific mendelian disease. C: to confirm and further reclassify into
specific subtypes. D: to confirm the clinical diagnoses. [3]I: treatment planning. II: further examination of the respiratory system. III: referral for systemic evaluation. IV: provision of appropriate genetic
counseling. [4]For P36, her mother carried a heterozygous c.1657+1G>T while his father did not. Further CNVs analysis based on ES data did not find any gross deletion or duplication of ITGB2 gene either in
the proband or her father. After excluding the possibility of non‐paternity based on an identity‐of‐descent analysis by using the trio‐ES data, we infer that there are two likely explanations: (i) a paternally
inherited deletion that is undetected by CNVkit based on ES data; (ii) maternal uniparental disomy.
HAO

Abbreviations: AD, autosomal dominant; AR, autosomal recessive; Hemi, hemizygous; Het, heterozygous; Homo, homozygous; LP, likely pathogenic; P, pathogenic; XLD, X‐linked dominant; XLR, X‐linked
recessive.
ET AL.
HAO ET AL. | 7

F I G U R E 2 Characteristics of inheritance patterns, distribution of the disease types, and clinical effects in 37 cases who received the
molecular diagnoses. (a) Proportion of inheritance patterns. (b) Distribution of the disease types. (c) Distribution of the effects of molecular
diagnoses by NGS‐based first‐tier testing. Category A, to produce new clinical diagnoses and guide the etiology. Category B, to change initial
diagnoses and identify specific mendelian disease. Category C, to confirm and further reclassify into specific subtypes. Category D, to confirm
the clinical diagnoses. AD, autosomal dominant; AR, autosomal recessive; ILD, interstitial lung disease; NGS, next‐generation sequencing;
PCD, primary ciliary dyskinesia; PIDs, primary immunodeficiency disorders; XLD, X linked dominant; XLR, X linked recessive

and the accumulation of knowledge of rare genetic diseases for Although phenotyping is a key step before performing genetic
clinicians. testing, a complete description of the patient's phenotypes is not
Patient 44, an 8‐month‐old boy, was referred to the clinic for easily accessible due to the clinical heterogeneity of the respiratory
recurrent fever. The patient presented mild to moderate anemia, diseases. Among these monogenic diseases, PIDs are genetically and
severe neutropenia, elevated erythrocyte sedimentation rate, and phenotypically heterogeneous disorders with over 300 candidate
C‐reactive protein. ES‐based testing identified pathogenic compound genes (Bousfiha et al., 2018). Furthermore, the correlation between
heterozygous variants in LPIN2, which led to the diagnosis of Majeed the phenotype and genotype in PIDs is complicated. PIDs are
syndrome (OMIM: 609628) (Liu et al., 2019). New subspecialist care grouped into 10 specific subtypes according to the International
and clinical management combined with hematology, even including Union of Immunological Societies (IUIS) PID classification (Bousfiha
additional diagnostic studies were initiated. et al., 2018). Patients with PIDs often require immediate clinical care
Patient 100, a 4‐month‐old boy, was referred to the clinic for to prevent recurrent severe infection, which may lead to fatality
tachypnea after birth. Chest HRCT (high‐resolution computed to- (Picard et al., 2018). The clinical variable phenotype of PIDs makes
mography) showed bilateral diffuse parenchyma infiltration. Mole- confirmative diagnosis challenging. Misdiagnosis may delay im-
cular testing found a pathogenic variant in SFTPC, which was a mediate and precise clinical care.
mutational hotspot and resulted in pulmonary surfactant metabolism PCD is another rare monogenic disease associated with cilia
dysfunction 2 (OMIM: 610913). The molecular findings helped clin- dysmotility and caused by over 40 causative genes. Children affected
icians to identify and confirm specific genetic causes of interstitial by PCD have progressive respiratory disease characterized by
lung disease, and provided a timely diagnosis by avoiding invasive bronchiectasis and impaired lung function (Horani et al., 2016;
lung biopsy. The timely diagnosis has allowed specific therapies and Mitchison & Valente, 2017). The clinical and genetic heterogeneity of
even referral for lung transplantation if indicated. PCD makes the diagnosis difficult, despite the availability of so-
phisticated diagnostic technologies, including nNO (nasal nitric
oxide), HSVA (high‐speed video‐microscopy analysis), TEM (trans-
4 | DI SCUSSION mission electron microscopy), IP (immunoprecipitation) of ciliary
proteins, and genetic testing. The “gold standard” to diagnose PCD
In the current study, we used exome sequencing (with or without was considered to be TEM detection. However, ultrastructural de-
CNV‐seq) for molecular diagnosis of inherited disease with re- fects were not shown for some patients with a genetic diagnosis of
spiratory system involvement and identified monogenic disorders of PCD, which may lead to misdiagnosis for these patients (Horani
37 patients (with an overall diagnostic rate of 34.6%). Our study et al., 2013; Schwabe et al., 2008). The American Thoracic Society
showed that the clinical heterogeneity and complex etiology of the has recently released Practice Guidelines on the diagnosis of primary
respiratory disorders made ES a valuable first‐tier diagnostic tool. ciliary dyskinesia (Shapiro et al., 2018). The Practice Guidelines
Of the positive cases, the two most frequent monogenic dis- propose a diagnostic algorithm for suspected PCD patients. It is re-
orders were PIDs and PCD, respectively, accounting for 73.0% commended that a panel of diagnostic tests is applied to diagnose
(27/37) of molecular diagnoses, which indicates the monogenic dis- PCD; however, considering the limitations of testing devices and
ease spectrum in the Department of Respiratory Medicine of Beijing local clinical expertise for TEM or nNO testing, genomic testing is a
Children's Hospital. cost‐effective and time‐efficient test. It improves the diagnostic
8 | HAO ET AL.

workflow, helping to diagnose PCD when TEM is not available and For undiagnosed samples, further investigation is required to be
confirming PCD in patients with inconclusive nNO or TEM results. taken into account: (i) to use RNA‐seq and genome sequencing to
The guidelines conditionally recommend using an extended genetic identify the putative pathogenic variants located in noncoding region
panel over TEM testing and/or standard (≤12 genes) genetic panel, (Cummings et al., 2017); (ii) to use a new algorithm to reanalyze CNV
but our practice results showed that exome sequencing tests were variants (Gross et al., 2019); (iii) to reanalyze ES data due to follow‐
more time‐efficient as sequencing cost is decreasing dramatically. up phenotypes and updated genetic disease databases (Ewans
Our results manifested that genomic testing could be applied as a et al., 2018); (iv) to identify genes with poor capture and low cov-
diagnostic alternative, especially in medical institutions with limited erage during ES due to special gene structure, and fill the gap with
medical resources in which the other diagnostic strategies, such as Sanger sequencing. According to our survey, the diagnostic yield of
TEM equipment or expertise, are not available (Rumman et al., 2017; patients aged 0–18 months and over 18 months was 29.1% (16/55)
Shoemark et al., 2017). Genetic testing not only provides a definitive and 40.4% (21/52), respectively. We speculate that the gradual
diagnosis but also extends beyond its diagnostic value, to the ap- evolution of the phenotypes over time increased the diagnostic yield
propriate genetic counseling of the affected individuals and their of genetic testing. The yield will be higher and genetic diagnostic
family members. Therefore, we prefer to perform exome sequencing results more successful if a more detailed clinical phenotype is
tests for suspected PCD, overcoming the pitfalls of other diagnostic available.
measures. To the best of our knowledge, this is the first single‐center study
In our cohort, the patients with the following clinical presenta- to investigate the genomic testing of inherited pediatric respiratory
tion are usually considered to be appropriate for genetic testing: (i) diseases in China. While a comprehensive and empirical clinical
family members with similar symptoms or suspected inheritance evaluation is critical, the power of ES‐based genomic testing is in-
diseases; (ii) specific or severe clinical phenotype; (iii) multiple de- dispensable in expediting the accurate diagnosis for genetic diseases,
formities; and (iv) exclusion of other factors (Boycott et al., 2015; particularly for those with heterogeneous and atypical clinical man-
Yao & Shen, 2017). Except for certain syndromes with respiratory ifestations. What's more, the increment of approximately 250 new
system involvement, the majority of rare monogenic diseases are Mendelian genes each year (Chong et al., 2015), the emergence of
often difficult to be recognized by respiratory clinicians due to the novel phenotypes, and the lack of accessibility to other testing
short‐term practice of medical genetics in China. According to our techniques all add further difficulties to the disease diagnosis. Our
study, 16.2% (6/37) of our diagnosed patients did not receive a study suggests that ES as a first‐tier test can rapidly identify mole-
clinical diagnosis before the genetic testing. The ES‐based genotype‐ cular defects and provide a helpful diagnostic strategy for further
testing approach retrospectively directed the clinical diagnosis individualized patient care and personalized genetic counseling in
without extensive clinical evaluation, enabling physicians to reach a pediatric respiratory medicine.
definite diagnosis.
One limitation of ES as a first‐tier test is that it increases the ACKNOWLEDGM E NTS
complexity of data analysis and interpretation compared with panel This study was partially supported by grants from the Ministry of
sequencing methods. These complexities are shown in the following Science and Technology of China (2016YFC1000306), the Beijing
aspects: (i) the interpretation of secondary findings; (ii) the eva- Municipal Science and Technology Commission Foundation
luation of de novo or null variants in the nonestablished disease (Z181100001918003), the Beijing Municipal Commission of Health
genes; (iii) the increased number and interpretation of variants of and Family Planning Foundation (2018‐2‐1141, 2020‐4‐1144), and
uncertain significance (VUS) in genes related to the primary phe- the Special Fund of the Pediatric Medical Coordinated Development
notype. Some measures have been taken to address the limitations. Center of Beijing Hospitals Authority (XTCX201807), Beihang
Genetic counseling was needed to inform probands and their par- University & Capital Medical University Advanced Innovation Center
ents and to address secondary findings and VUS variants. For the for Big Data‐Based Precision Medicine Plan (BHME‐201905).
evaluation of genes of unknown significance, especially for sus-
pected candidate genes, functional studies are required, although CO N FLI CT O F I N TER E S TS
these are time‐consuming and are unable to provide a timely mo- The authors declare that there are no conflict of interests.
lecular diagnosis. To date, we have found several suspected can-
didate genes whose pathological mechanisms were not yet clear. A U T H OR C O N T R I B U T I ON S
Functional studies are ongoing in our laboratory. The limitations of Chanjuan Hao, Baoping Xu, and Wei Li conceived and designed the
ES also include technical issues affecting sequencing depth and study. Jun Liu, Yao Yao, Zhipeng Zhao, Jun Yin, Lanqin Chen, and Hao
coverage, and bioinformatics challenges to identify CNVs based on Wang recruited their respective patients to this study and provided
ES data. In future work, we will define genes with inherited pul- clinical data regarding their patients. Ruolan Guo, Xuyun Hu, Jun
monary conditions that are not well captured or sequenced at a low Guo, and Zhan Qi contributed to the analysis of sequencing data.
depth and further optimize the genomic sequencing pipeline, to Chanjuan Hao, Ruolan Guo, and Jun Liu were involved in manuscript
reach a higher diagnostic rate. editing. All authors reviewed and approved the final version.
HAO ET AL. | 9

W E B R E S O UR C E S in Mendelian disease with transcriptome sequencing. Science

Bioinformatics databases and tools utilized for exome sequencing Translational Medicine, 9(386), eaal5209. https://doi.org/10.1126/
scitranslmed.aal5209
interpretation.
Damseh, N., Quercia, N., Rumman, N., Dell, S. D., & Kim, R. H. (2017).
Online Mendelian Inheritance in Man, OMIM. URL: https:// Primary ciliary dyskinesia: mechanisms and management. The
omim.org/ Application of Clinical Genetics, 10, 67–74. https://doi.org/10.2147/
ClinVar. URL: https://www.ncbi.nlm.nih.gov/clinvar/ TACG.S127129
Ewans, L. J., Schofield, D., Shrestha, R., Zhu, Y., Gayevskiy, V., Ying, K., &
Human Gene Mutation Database, HGMD. URL: https://portal.
Roscioli, T. (2018). Whole‐exome sequencing reanalysis at
biobase-international.com/hgmd/ 12 months boosts diagnosis and is cost‐effective when applied
MalaCards: The human disease database. URL: https://www. early in Mendelian disorders. Genetics in Medicine, 20(12),
malacards.org/ 1564–1574. https://doi.org/10.1038/gim.2018.39
GnomAD Browser. URL: http://gnomad.broadinstitute.org/ Fassad, M. R., Patel, M. P., Shoemark, A., Cullup, T., Hayward, J., Dixon, M.,
& Mitchison, H. M. (2020). Clinical utility of NGS diagnosis and
1000 Genomes. URL: https://www.internationalgenome.org/
disease stratification in a multiethnic primary ciliary dyskinesia
dbSNP. URL: https://www.ncbi.nlm.nih.gov/snp/ cohort. Journal of Medical Genetics, 57(5), 322–330. https://doi.org/
PolyPhen‐2. URL: http://genetics.bwh.harvard.edu/pph2 10.1136/jmedgenet-2019-106501
SIFT. URL: http://sift.bii.a-star.edu.sg/ Gross, A. M., Ajay, S. S., Rajan, V., Brown, C., Bluske, K., Burns, N. J., &
Taft, R. J. (2019). Copy‐number variants in clinical genome
CADD. URL: https://cadd.gs.washington.edu/
sequencing: deployment and interpretation for rare and
MutationTaster. URL: http://www.mutationtaster.org/ undiagnosed disease. Genetics in Medicine, 21(5), 1121–1130.
TGex. URL: https://geneyx.com/geneyxanalysis/ https://doi.org/10.1038/s41436-018-0295-y
Horani, A., Brody, S. L., Ferkol, T. W., Shoseyov, D., Wasserman, M. G., Ta‐
DATA AV AILA BILITY STATEMENT shma, A., & Kerem, E. (2013). CCDC65 mutation causes primary
ciliary dyskinesia with normal ultrastructure and hyperkinetic cilia.
The data that support the findings of this study are available from
PLoS One, 8(8), e72299. https://doi.org/10.1371/journal.pone.
the corresponding author upon reasonable request. All identified 0072299
pathogenic, likely pathogenic variants were submitted to ClinVar Horani, A., Ferkol, T. W., Dutcher, S. K., & Brody, S. L. (2016). Genetics and
(https://www.ncbi.nlm.nih.gov/clinvar/). The data are publicly acces- biology of primary ciliary dyskinesia. Paediatric Respiratory Reviews,
18, 18–24. https://doi.org/10.1016/j.prrv.2015.09.001
sible under the accession number SCV001499898 ‐ SCV001499945.
Hu, X., Guo, R., Guo, J., Qi, Z., Li, W., & Hao, C. (2020). Parallel tests of
whole exome sequencing and copy number variant sequencing
ORCID increase the diagnosis yields of rare pediatric disorders. Frontiers in
Ruolan Guo http://orcid.org/0000-0002-3378-2008 Genetics, 11, 473. https://doi.org/10.3389/fgene.2020.00473
Hu, X., Li, N., Xu, Y., Li, G., Yu, T., Yao, R. E., & Shen, Y. (2018). Proband‐
Jun Guo http://orcid.org/0000-0002-3685-939X
only medical exome sequencing as a cost‐effective first‐tier
genetic diagnostic test for patients without prior molecular tests
R EF E RE N C E S and clinical diagnosis in a developing country: The China
Arts, P., Simons, A., AlZahrani, M. S., Yilmaz, E., AlIdrissi, E., experience. Genetics in Medicine, 20(9), 1045–1053. https://doi.
van Aerde, K. J., & Hoischen, A. (2019). Exome sequencing in routine org/10.1038/gim.2017.195
diagnostics: A generic test for 254 patients with primary Hu, X., Liu, J., Guo, R., Guo, J., Zhao, Z., Li, W., & Hao, C. (2019). A novel
immunodeficiencies. Genome Medicine, 11(1), 38. https://doi.org/10. 14q13.1‐21.1 deletion identified by CNV‐Seq in a patient with
1186/s13073-019-0649-3 brain‐lung‐thyroid syndrome, tooth agenesis and immunodeficiency.
Bousfiha, A., Jeddane, L., Picard, C., Ailal, F., Bobby Gaspar, H., Al‐Herz, Molecular Cytogenetics, 12, 51. https://doi.org/10.1186/s13039-019-
W., & Sullivan, K. E. (2018). The 2017 IUIS phenotypic classification 0463-z
for primary immunodeficiencies. Journal of Clinical Immunology, Kearney, H. M., Thorland, E. C., Brown, K. K., Quintero‐Rivera, F., &
38(1), 129–143. https://doi.org/10.1007/s10875-017-0465-8 South, S. T., Working Group of the American College of Medical
Boycott, K., Hartley, T., Adam, S., Bernier, F., Chong, K., Fernandez, B. A., & Genetics Laboratory Quality Assurance, C. (2011). American
Canadian College of Medical, G. (2015). The clinical application of College of Medical Genetics standards and guidelines for
genome‐wide sequencing for monogenic diseases in Canada: interpretation and reporting of postnatal constitutional copy
Position Statement of the Canadian College of Medical number variants. Genetics in Medicine, 13(7), 680–685. https://doi.
Geneticists. Journal of Medical Genetics, 52(7), 431–437. https:// org/10.1097/GIM.0b013e3182217a3a
doi.org/10.1136/jmedgenet-2015-103144 Liu, J., Hu, X. Y., Zhao, Z. P., Guo, R. L., Guo, J., Li, W., & Xu, B. P. (2019).
Chong, J. X., Buckingham, K. J., Jhangiani, S. N., Boehm, C., Sobreira, N., Compound heterozygous LPIN2 pathogenic variants in a patient
Smith, J. D., & Bamshad, M. J. (2015). The genetic basis of mendelian with Majeed syndrome with recurrent fever and severe
phenotypes: Discoveries, challenges, and opportunities. American neutropenia: Case report. BMC Medical Genetics, 20(1), 182.
Journal of Human Genetics, 97(2), 199–215. https://doi.org/10.1016/ https://doi.org/10.1186/s12881-019-0919-3
j.ajhg.2015.06.009 Marshall, C. R., Scherer, S. W., Zariwala, M. A., Lau, L., Paton, T. A.,
Cifaldi, C., Brigida, I., Barzaghi, F., Zoccolillo, M., Ferradini, V., Stockley, T., & Kim, R. H. (2015). Whole‐exome sequencing and
Petricone, D., & Di Matteo, G. (2019). Targeted NGS platforms for targeted copy number analysis in primary ciliary dyskinesia. G3
genetic screening and gene discovery in primary (Bethesda), 5(8), 1775–1781. https://doi.org/10.1534/g3.115.
immunodeficiencies. Frontiers in Immunology, 10, 316. https://doi. 019851
org/10.3389/fimmu.2019.00316 Meng, L., Pammi, M., Saronwala, A., Magoulas, P., Ghazi, A. R., Vetrini, F., &
Cummings, B. B., Marshall, J. L., Tukiainen, T., Lek, M., Donkervoort, S., Lalani, S. R. (2017). Use of exome sequencing for infants in intensive
Foley, A. R., & MacArthur, D. G. (2017). Improving genetic diagnosis care units: Ascertainment of severe single‐gene disorders and effect
10 | HAO ET AL.

on medical management. JAMA Pediatrics, 171(12), e173438. DNAH11 mutations. Human Mutation, 29(2), 289–298. https://doi.
https://doi.org/10.1001/jamapediatrics.2017.3438 org/10.1002/humu.20656
Mitchison, H. M., & Valente, E. M. (2017). Motile and non‐motile cilia in Scocchia, A., Wigby, K. M., Masser‐Frye, D., Del Campo, M.,
human pathology: From function to phenotypes. Journal of Galarreta, C. I., Thorpe, E., & Taft, R. J. (2019). Clinical whole
Pathology, 241(2), 294–309. https://doi.org/10.1002/path.4843 genome sequencing as a first‐tier test at a resource‐limited
Orloff, M. S., He, X., Peterson, C., Chen, F., Chen, J. L., Mester, J. L., & dysmorphology clinic in Mexico. NPJ Genomic Medicine, 4, 5.
Eng, C. (2013). Germline PIK3CA and AKT1 mutations in Cowden https://doi.org/10.1038/s41525-018-0076-1
and Cowden‐like syndromes. American Journal of Human Genetics, Shapiro, A. J., Davis, S. D., Polineni, D., Manion, M., Rosenfeld, M., &
92(1), 76–80. https://doi.org/10.1016/j.ajhg.2012.10.021 Dell, S. D., American Thoracic Society Assembly on Pediatrics.
Picard, C., Bobby Gaspar, H., Al‐Herz, W., Bousfiha, A., Casanova, J. L., (2018). Diagnosis of primary ciliary dyskinesia. An Official American
Chatila, T., & Sullivan, K. E. (2018). International Union of Immunological Thoracic Society Clinical Practice Guideline. American Journal of
Societies: 2017 primary immunodeficiency diseases committee report Respiratory and Critical Care Medicine, 197(12), e24–e39. https://doi.
on inborn errors of immunity. Journal of Clinical Immunology, 38(1), org/10.1164/rccm.201805-0819ST
96–128. https://doi.org/10.1007/s10875-017-0464-9 Shoemark, A., Frost, E., Dixon, M., Ollosson, S., Kilpin, K., Patel, M., &
Richards, S., Aziz, N., Bale, S., Bick, D., Das, S., Gastier‐Foster, J., & Hogg, C. (2017). Accuracy of immunofluorescence in the diagnosis
Committee, A. L. Q. A. (2015). Standards and guidelines for the of primary ciliary dyskinesia. American Journal of Respiratory and
interpretation of sequence variants: A joint consensus Critical Care Medicine, 196(1), 94–101. https://doi.org/10.1164/rccm.
recommendation of the American College of Medical Genetics and 201607-1351OC
Genomics and the Association for Molecular Pathology. Genetics in Stavropoulos, D. J., Merico, D., Jobling, R., Bowdin, S., Monfared, N.,
Medicine, 17(5), 405–424. https://doi.org/10.1038/gim.2015.30 Thiruvahindrapuram, B., & Marshall, C. R. (2016). Whole genome
Riggs, E. R., Andersen, E. F., Cherry, A. M., Kantarci, S., Kearney, H., & sequencing expands diagnostic utility and improves clinical
Patel, A., ACMG. (2020). Technical standards for the interpretation management in pediatric medicine. NPJ Genomic Medicine, 1,
and reporting of constitutional copy‐number variants: A joint 15012. https://doi.org/10.1038/npjgenmed.2015.12
consensus recommendation of the American College of Medical Yang, Y., Muzny, D. M., Reid, J. G., Bainbridge, M. N., Willis, A., Ward, P. A.,
Genetics and Genomics (ACMG) and the Clinical Genome Resource & Eng, C. M. (2013). Clinical whole‐exome sequencing for the
(ClinGen). Genetics in Medicine, 22(2), 245–257. https://doi.org/10. diagnosis of mendelian disorders. New England Journal of Medicine,
1038/s41436-019-0686-8 369(16), 1502–1511. https://doi.org/10.1056/NEJMoa1306555
Rudilla, F., Franco‐Jarava, C., Martinez‐Gallo, M., Garcia‐Prat, M., Martin‐ Yao, Y., & Shen, K. J. (2017). Monogenic diseases in respiratory medicine:
Nalda, A., Riviere, J., & Colobran, R. (2019). Expanding the clinical Clinical perspectives. Pediatric Investigation, 1(1), 27–31.
and genetic spectra of primary immunodeficiency‐related disorders
with clinical exome sequencing: expected and unexpected findings.
Frontiers in Immunology, 10, 2325. https://doi.org/10.3389/fimmu. SU PP O RT I N G I N FO RM ATI ON
2019.02325 Additional Supporting Information may be found online in the
Rumman, N., Jackson, C., Collins, S., Goggin, P., Coles, J., & Lucas, J. S.
supporting information tab for this article.
(2017). Diagnosis of primary ciliary dyskinesia: potential options for
resource‐limited countries. European Respiratory Review, 26(143),
160058. https://doi.org/10.1183/16000617.0058-2016
Sanford, E. F., Clark, M. M., Farnaes, L., Williams, M. R., Perry, J. C.,
How to cite this article: Hao, C., Guo, R., Liu, J., Hu, X., Guo, J.,
Ingulli, E. G., & Investigators, R. (2019). Rapid whole genome
sequencing has clinical utility in children in the PICU. Pediatric Yao, Y., Zhao, Z., Qi, Z., Yin, J., Chen, L., Wang, H., Xu, B., & Li,
Critical Care Medicine, 20(11), 1007–1020. https://doi.org/10.1097/ W. (2021). Exome sequencing as the first‐tier test for
PCC.0000000000002056 pediatric respiratory diseases: A single‐center study. Human
Schwabe, G. C., Hoffmann, K., Loges, N. T., Birker, D., Rossier, C.,
Mutation, 1–10. https://doi.org/10.1002/humu.24216
de Santi, M. M., & Bartoloni, L. (2008). Primary ciliary dyskinesia
associated with normal axoneme ultrastructure is caused by