Progress in the application of metagenomic next-generation sequencing in pediatric infectious diseases

Abstract

Infectious diseases are the major cause of children’s deaths all over the world. With the development of evidence-based medicine, etiological diagnosis becomes more and more important. Since traditional methods have been unable to meet the needs of diagnosis and treatment, metagenomic next-generation sequencing (mNGS) gradually shows its unique advantages for pathogen diagnosis. This article aimed to introduce the application of mNGS technology in the diagnosis and treatment of neonatal and puerile infectious diseases by providing some examples. Infectious diseases are the major cause of children’s deaths all over the world. With the development of evidence-based medicine, etiological diagnosis becomes more and more important. Since traditional methods have been unable to meet the needs of diagnosis and treatment, metagenomic next-generation sequencing (mNGS) gradually shows its unique advantages for pathogen diagnosis. This article aimed to introduce the application of mNGS technology in the diagnosis and treatment of neonatal and puerile infectious diseases by providing some examples. 1. IntroductionGlobally, 6.3 million children died before the age of 5 years in 2013, and 51.8% (3.257 million) of these deaths were due to infectious diseases.1Liu L. Oza S. Hogan D. Perin J. Rudan L. Lawn J.E. et al.Global, regional, and national causes of child mortality in 2000-13, with projections to inform post-2015 priorities: an updated systematic analysis.Lancet. 2015; 385: 430-440Abstract Full Text Full Text PDF PubMed Scopus (1920) Google Scholar Pneumonia, diarrhea, and measles were collectively responsible for half of these deaths. Infectious disease is also the major cause of neonatal deaths worldwide, accounting for 44% (2.761 million) of all pediatric deaths.2World Health Statistics 2019 Monitoring health for the SDGs. World Health Organization, Geneva, Switzerland2019Google Scholar Due to the occult onset, atypical clinical signs, and rapid progression, neonatal infectious diseases that are associated with high mortality rates are difficult to diagnose. According to the statistics of the World Health Organization in 2018,3World Health Statistics 2018 Monitoring health for the SDGs. World Health Organization, Geneva, Switzerland2018Google Scholar prematurity, intrapartum-related events, and neonatal sepsis accounted for almost three-quarters of all neonatal deaths, which mostly occurred in the first week of life. Oza et al.4Oza S. Lawn J.E. Hogan D.R. Mathers C. Cousens S.N. Neonatal cause-of-death estimates for the early and late neonatal periods for 194 countries: 2000–2013.Bull World Health Organ. 2015; 93: 19-28Crossref PubMed Scopus (205) Google Scholar found that infections caused nearly half of late neonatal deaths in 194 countries between 2000 and 2013.Traditional methods to identify pathogens of infections include bacterial culture, enzyme-linked immunosorbent assay (ELISA), polymerase chain reaction (PCR), and others.5Xue H.M. Lu X.M. Li W. Progress of detection methods in the diagnosis of infectious disease.Chin J Obstet Gynecol Pediatr. 2012; 8 (https://doi.org/10.3877/cma.j.issn.1673-5250.2012.04.037): 441-443Google Scholar However, all of these methods are affected by multiple factors such as the use of antibiotics, the amount of the specimen, and the populations of pathogen, which could result in false-negative results. In addition, these traditional methods cannot identify multiple pathogens in a single detection, and multiple sampling might result in iatrogenic anemia and prolong the duration of hospitalization. Thus, traditional methods are not efficient for the detection of pathogens in cases of infectious disease. These disadvantages are particularly relevant for diagnosis in infants and children, as infections in these populations are more difficult to diagnose due to atypical clinical signs.In recent years, the development of metagenomic next-generation sequencing (mNGS) has shown unique technical advantages and wide application prospects for detecting and analyzing pathogens of infectious disease, especially those that easily occur in children. With mNGS, it is possible to directly extract genomic DNA from samples comprising several microbial species without the need for isolation and culturing, which is particularly useful in cases where the identities of the pathogen are unknown. Furthermore, traditional methods require isolation and culture in vitro, and many pathogenic microorganisms in the environment or human body cannot readily be cultured, and thus may only be identified by mNGS. The overall efficiency and positivity rate of traditional detection methods are low, and PCR detection is limited to amplifying the gene sequences of known pathogenic microorganisms. By contrast, mNGS can accurately identify pathogens using high-throughput sequencing, which is a particular advantage for detecting unknown viruses. For example, in 2019, a previously unknown beta-coronavirus was discovered in samples from patients with pneumonia in Wuhan, China, through the use of this unbiased sequencing method.6Zhu N. Zhang D. Wang W. Li X. Yang B. Song J. et al.A novel coronavirus from patients with pneumonia in China, 2019.N Engl J Med. 2020; 382: 727-733Crossref PubMed Scopus (14761) Google Scholar This novel coronavirus was ultimately named SARS-CoV-2 as the cause of the global COVID-19 pandemic.In this review, I focus on the application of mNGS technology reported to date, highlighting its important role and value in the diagnosis and treatment of neonatal and pediatric infectious disease. This review should serve as a useful guide for pediatricians and clinical microbiologists to become familiar with this new tool and understand its advantages and limitations.2. General background of mNGS technologyInitially, genome sequencing involved sequencing by synthesis (Sanger sequencing) and sequencing by cleavage (Maxam-Gilbert sequencing), which was first established in 1977.7Ambardar S. Gupta R. Trakroo D. Lal R. Vakhlu J. High throughput sequencing: an overview of sequencing chemistry.Indian J Microbiol. 2016; 56: 394-404Crossref PubMed Scopus (95) Google Scholar Sanger sequencing then developed into first-generation sequencing, which was used to obtain the first human genome map in 2001. Since then, various techniques have been developed that are collectively known as “next-generation sequencing” (NGS), which can be further classified into second-generation sequencing, characterized by parallel sequencing of massive DNA molecules, and third-generation sequencing, characterized by single-molecule sequencing. The most common applications of NGS for diagnosis include: (1) targeted amplicon sequencing, which involves enriching specific gene regions with different methods, followed by NGS; (2) whole-genome sequencing, which involves sequencing and assembly of the genome of a pathogen of interest; and (3) mNGS, which involves sequencing all of the nucleic acids (DNA and/or RNA) of both the host and pathogen(s) of a specimen in parallel.8Simner P.J. Miller S. Carroll K.C. Understanding the promises and hurdles of metagenomic next-generation sequencing as a diagnostic tool for infectious diseases.Clin Infect Dis. 2018; 66 (778–88)Crossref PubMed Scopus (237) Google Scholar These three main applications of NGS are summarized in Table 1. This review focuses on the application of mNGS in diagnosing infectious diseases in infants and children.Table 1Comparison of next-generation sequencing (NGS) applications in clinical microbiology.ApplicationApproachAdvantagesDisadvantagesTargeted amplicon NGSSequencing of specifically enriched targets using NGS•Faster, less complicated, and cheaper•Highly sensitive•The most commonly used microbiota profiling strategy•Eliminates unwanted sequences•Helpful where slow-growing pathogens are suspected•Often lacks the discriminatory power to differentiate prokaryotes at the species taxonomic level•The accuracy depends on the quality and completeness of the reference databases used•Biased•Additional cost for enrichment stepsWhole-genome sequencing (WGS)Sequencing and assembly of DNA fragments from across an entire microbial genome•Especially useful in hospital and public health epidemiologic studies•Provides potentially untapped information regarding pathogen virulence•Predicts antimicrobial resistance•Provides entire genome and plasmid sequences•Allows for single-nucleotide polymorphism and mutational analysis•The presence of other clinically insignificant bacteria in specimens may confound WGS results•Cannot accurately predict susceptibility•May be inadequate in detecting mechanisms of resistance that have not been fully elucidatedMetagenomic NGSSequencing all nucleic acids of both the host and microbial community within a specimen in parallel•Provides unbiased pathogen detection directly from patient specimens•Not only identifies microorganisms per se, but may also provide information on the types of (microbial) genes present•Provides very high sensitivity for the diagnostics of rare pathogens•Offers a much broader description of microbial community genetics•Ability to strain type directly from specimens, study virulence, antimicrobial resistance genes, and evaluate the host immune response•High price of technologies and machines•Long turn-around times•The majority of reads are of human origin•Requires specific expertise in performing, testing, and analyzing data•Accuracy relies on the availability and completeness of specific reference databases•Lack of standardization and automation Open table in a new tab 3. Applications of mNGS in infectious disease diagnosisIn 2014, a 14-year-old male patient with severe combined immunodeficiency to a three over a of with specific S.N. et diagnosis of by next-generation Engl J Med. PubMed Scopus Google Scholar unbiased NGS of the which been in traditional clinical This was the first in which the clinical of mNGS was mNGS can be into nucleic sequencing, and of the detection by mNGS not on bacterial the sequences of pathogens are directly from the samples and then the sequences are and using By unbiased sequencing, rapid and reference mNGS can be for pathogen of mNGS for the detection of are the most common pathogens infectious diseases in and children. is the for diagnosis. However, not only culture require a but it can also be by a of factors such as specimen and culture and the and of the which could to of bacteria in the etiological of an in a child is and culture, second-generation sequencing is as the detection et N. S. et detection of pathogens in children with infections by next-generation 2018; PubMed Scopus Google Scholar or samples from children were with and samples from children were with by reads per million reads and of the reads of the bacteria were as the The sequencing reads of bacteria in culture were identified by NGS at all of patients and were to the bacteria in 8 of bacteria were by NGS in patients with before disease a of resistance genes were in all samples by NGS is a new to identify pathogens in and may predict in some et Yang J. J. and methods of diagnosis and of pediatric 2020; Full Text Full Text PDF PubMed Scopus (3) Google Scholar that the of a pathogen in pediatric patients was The rate of infections was and in some cases it was as high as that mNGS technology can not only identify the pathogens patients with but can also the genes responsible for pathogen advantage of NGS is that only of is for which can thus the of pediatric In contrast, traditional methods such as culturing, and require at several of for However, this amount of is almost to obtain of the in pediatric et et diagnostic value of metagenomic next-generation sequencing for in bacterial Infect Dis. PubMed Scopus Google Scholar cases that were with at from to The that patients for S. in and/or and in mNGS. samples were identified as S. only using mNGS mNGS and culture, which was previously as the for the diagnosis of bacterial the sensitivity and of S. by mNGS were and and the value and value were and the that mNGS shows high sensitivity and for the diagnosis of S. also found that mNGS was less affected by the use of before et C. N. D. S. R. et due to bacteria in a Infect Full Text Full Text PDF PubMed Scopus (3) Google Scholar reported the of a with and and a diagnosis of was on a culture of at and at and culture of at also metagenomic sequencing of the which S. and This that mNGS is in bacteria that cannot easily be by et Wang The of and pneumonia in very J 2020; Scholar infants to the neonatal and to the or the to or not were with samples on and were by high-throughput sequencing, that the of in the was that in the on which was the of before the of the of in the of infants Thus, in the may to clinical infections caused by of mNGS for detecting and PCR are the main methods for detecting it is difficult to obtain an early diagnosis using are only several and it is also difficult to obtain an and diagnosis using PCR of the need to predict the type of in mNGS has to these For children with of onset, or and rapid disease in the of high multiple PCR detection and culture of the can first be such as PCR of the to a this is NGS should then be However, it should be that the nucleic acids of a are before mNGS, the should be into et L. Wang Wang et genomic next-generation sequencing of a of with and J Pediatr. 2020; Scholar reported a with and and multiple of and were for mNGS the presence of in specimens of the the diagnosis of treatment with for three and was from the et J. R. S. S. et of in cases of pediatric and using 2016; PubMed Scopus Google Scholar the and samples of pediatric patients with of unknown identified specific pathogens in three of samples and in of This study that mNGS is an unbiased and that is useful for the detection of in pediatric patients with et W. of in samples from infants and children with in using J. 2020; PubMed Scopus Google Scholar used mNGS to in samples from infants and children with and of from to in found that human which was in of was the cause of followed by an of a of human and other viruses. mNGS discovered such as and These samples were also for the presence of using a PCR However, this could not identify many that were using the mNGS This study the of the mNGS to in samples of infants and children with in et N. Li Lu R. Wang et of in samples from children with severe in 2016; Full Text Full Text PDF Scopus Google Scholar from children with and without severe in between and These samples were to metagenomic using an NGS The that of the and were the most identified in the of children with and of the were also found in the of children without the mNGS this study the first of the in the of both children with and without which can be of for the global and of an advantage in the rapid of novel pathogens to its of all types of For example, in the recent of et Wang L. et of a novel coronavirus severe pneumonia in a J 2020; PubMed Scopus Google Scholar clinical and specimens from patients with severe pneumonia from in 2019, the nucleic and using the mNGS The sequencing the presence of a previously unknown beta-coronavirus strain in a in all which with the of Thus, the mNGS can also to predict the and and other X. Applications and of metagenomic next-generation sequencing in the detection of 2019 novel J Med. 2020; of mNGS in infections have been gradually in recent years, such as and These infections can be caused by several pathogens Thus, an efficient and to identify such pathogens in infections be very important and for clinical et Yang L. B. et metagenomic sequencing with sequencing to clinical in with a Infect Microbiol. PubMed Scopus Google Scholar with a and that mNGS a positivity rate the samples a result by mNGS, the pathogen could only be in 5 samples by In the microbial species were by mNGS, bacteria and three that mNGS can more pathogen mNGS also advantage in the to pathogens in been with the patients that to pathogens could be in by mNGS with only using Furthermore, patient was with of both and and patient was with both and This study the advantages of mNGS for detecting multiple infections in a single et of clinical applications of detection of pathogenic microbial in in the treatment of severe pneumonia in Med. Scholar a to the application value of mNGS in the treatment of severe pneumonia in children. from children with severe pneumonia to the pediatric of from 2018 to and the samples for the detection of pathogenic by mNGS. the mNGS cases of cases of bacterial cases of bacterial of cases of and cases of these were patients with with the methods, cases of cases of bacterial cases of bacterial of cases of and cases of a of were identified by mNGS. mNGS a detection rate the traditional The treatment were for patients to the of mNGS, and clinical of patients this that mNGS has a advantage in detecting multiple pathogens and of mNGS in other can also be used for the diagnosis of pathogens other bacteria and viruses. infants with a and to a hospital in in the of S. Li J. sequencing in Scopus Google Scholar patients a in the of in the or with those in with rapid diagnostic in the samples of the first patient and in the of the mNGS of DNA from the identified and DNA reads from the genome in the samples of the DNA sequences from the genome were not in the on the of and the unique DNA sequences of these patients were with and were with and a treatment, the of and in the and to in both patients without of of mNGS in other from detecting mNGS is also useful in many other is a cause of and mortality in but the of has not been in the the early of and infants using mNGS and found that detection of the of and as an in et sequencing with in and mortality in 2016; Full Text Full Text PDF PubMed Scopus Google Scholar Furthermore, this study that the early of pathogens and resistance using mNGS can to and an of a caused by in a C. Li X. C. J. et in a neonatal new from next-generation sequencing Microbiol. 2018; PubMed Scopus Google Scholar were from patients and three from the sequencing was then used to the bacterial and evaluate the of S. in from different sampling The that multiple could be involved in S. in In addition, genes associated with resistance were identified in all that sequencing not only become for but also has for the of of S. This study also found that the of S. was in all samples in the that the used to of the of a patient was study in metagenomic sequencing to the microbiota of at high of a and the of a by with and the of several different species such as B. B. and with C. D. S. et sequencing of the both a species and of 2020; PubMed Scopus Google Scholar Health and the for and of the used mNGS for the of diseases such as diseases caused by and D.R. J. et in public Engl J Med. PubMed Scopus Google important use of mNGS is the of bacterial mNGS is used for detecting and the presence of pathogens in and then to the D.R. J. et in public Engl J Med. PubMed Scopus Google Scholar Furthermore, the of by mNGS for the and development of new diagnostics and of mNGS may provide an unbiased of DNA or sequences in a single and has the advantages of price and high it has some cause in genome which may result in a of sequencing in human to a PubMed Scopus Google Scholar mNGS cannot directly and This the of to the to this is in limited by as as sequencing and of a of sequences by mNGS to the which may result in a of pathogens and the sensitivity of pathogen B. of metagenomic sequencing in detecting the pathogens of infectious 2018; Scholar However, pathogen reads are more reads can be to more provide a diagnosis and for advantage of sequencing host nucleic is the to evaluate the host associated with the presence and type of P.J. Miller S. Carroll K.C. Understanding the promises and hurdles of metagenomic next-generation sequencing as a diagnostic tool for infectious diseases.Clin Infect Dis. 2018; 66 (778–88)Crossref PubMed Scopus (237) Google DNA can be of its and the microbial species and by sequencing may be most of the on the presence of a which could to due to different S. Gupta R. Trakroo D. Lal R. Vakhlu J. High throughput sequencing: an overview of sequencing chemistry.Indian J Microbiol. 2016; 56: 394-404Crossref PubMed Scopus (95) Google it is difficult to bacteria and pathogenic bacteria of samples from the and have aimed to pathogens from pathogens for many recent is the use of to the microbial as a to reads in to the bacterial in of J. D. P.J. et for in microbial by 2016; PubMed Scopus Google the of a is a major of sequencing for pathogenic diagnosis in infectious disease.Chin J Infect Dis. Scholar and the of and for mNGS cannot meet the need for rapid mNGS not on traditional methods, but it directly sequences nucleic acids in clinical samples with high mNGS can and a of pathogenic and in clinical which is especially in cases of and severe and With the of mNGS technology and the in clinical mNGS is used in clinical 1. IntroductionGlobally, 6.3 million children died before the age of 5 years in 2013, and 51.8% (3.257 million) of these deaths were due to infectious diseases.1Liu L. Oza S. Hogan D. Perin J. Rudan L. Lawn J.E. et al.Global, regional, and national causes of child mortality in 2000-13, with projections to inform post-2015 priorities: an updated systematic analysis.Lancet. 2015; 385: 430-440Abstract Full Text Full Text PDF PubMed Scopus (1920) Google Scholar Pneumonia, diarrhea, and measles were collectively responsible for half of these deaths. Infectious disease is also the major cause of neonatal deaths worldwide, accounting for 44% (2.761 million) of all pediatric deaths.2World Health Statistics 2019 Monitoring health for the SDGs. World Health Organization, Geneva, Switzerland2019Google Scholar Due to the occult onset, atypical clinical signs, and rapid progression, neonatal infectious diseases that are associated with high mortality rates are difficult to diagnose. According to the statistics of the World Health Organization in 2018,3World Health Statistics 2018 Monitoring health for the SDGs. World Health Organization, Geneva, Switzerland2018Google Scholar prematurity, intrapartum-related events, and neonatal sepsis accounted for almost three-quarters of all neonatal deaths, which mostly occurred in the first week of life. Oza et al.4Oza S. Lawn J.E. Hogan D.R. Mathers C. Cousens S.N. Neonatal cause-of-death estimates for the early and late neonatal periods for 194 countries: 2000–2013.Bull World Health Organ. 2015; 93: 19-28Crossref PubMed Scopus (205) Google Scholar found that infections caused nearly half of late neonatal deaths in 194 countries between 2000 and 2013.Traditional methods to identify pathogens of infections include bacterial culture, enzyme-linked immunosorbent assay (ELISA), polymerase chain reaction (PCR), and others.5Xue H.M. Lu X.M. Li W. Progress of detection methods in the diagnosis of infectious disease.Chin J Obstet Gynecol Pediatr. 2012; 8 (https://doi.org/10.3877/cma.j.issn.1673-5250.2012.04.037): 441-443Google Scholar However, all of these methods are affected by multiple factors such as the use of antibiotics, the amount of the specimen, and the populations of pathogen, which could result in false-negative results. In addition, these traditional methods cannot identify multiple pathogens in a single detection, and multiple sampling might result in iatrogenic anemia and prolong the duration of hospitalization. Thus, traditional methods are not efficient for the detection of pathogens in cases of infectious disease. These disadvantages are particularly relevant for diagnosis in infants and children, as infections in these populations are more difficult to diagnose due to atypical clinical signs.In recent years, the development of metagenomic next-generation sequencing (mNGS) has shown unique technical advantages and wide application prospects for detecting and analyzing pathogens of infectious disease, especially those that easily occur in children. With mNGS, it is possible to directly extract genomic DNA from samples comprising several microbial species without the need for isolation and culturing, which is particularly useful in cases where the identities of the pathogen are unknown. Furthermore, traditional methods require isolation and culture in vitro, and many pathogenic microorganisms in the environment or human body cannot readily be cultured, and thus may only be identified by mNGS. The overall efficiency and positivity rate of traditional detection methods are low, and PCR detection is limited to amplifying the gene sequences of known pathogenic microorganisms. By contrast, mNGS can accurately identify pathogens using high-throughput sequencing, which is a particular advantage for detecting unknown viruses. For example, in 2019, a previously unknown beta-coronavirus was discovered in samples from patients with pneumonia in Wuhan, China, through the use of this unbiased sequencing method.6Zhu N. Zhang D. Wang W. Li X. Yang B. Song J. et al.A novel coronavirus from patients with pneumonia in China, 2019.N Engl J Med. 2020; 382: 727-733Crossref PubMed Scopus (14761) Google Scholar This novel coronavirus was ultimately named SARS-CoV-2 as the cause of the global COVID-19 pandemic.In this review, I focus on the application of mNGS technology reported to date, highlighting its important role and value in the diagnosis and treatment of neonatal and pediatric infectious disease. This review should serve as a useful guide for pediatricians and clinical microbiologists to become familiar with this new tool and understand its advantages and

Topics & Concepts