Next-generation Sequencing of Cerebrospinal Fluid for the Diagnosis of Unexplained Central Nervous System Infections

Abstract

Background

Central nervous system infections cause substantial morbidity and mortality in pediatric patients. However, in approximately half of the clinical cases, the etiology is unidentified. As an unbiased molecular diagnostic technology, next-generation sequencing is gradually being applied to investigate central nervous system infections. This review summarizes and critiques the literature on this new technology for etiologic identification of unexplained central nervous system infections in pediatric patients and discusses the future prospects for development of this technology in pediatrics.

Methods

A comprehensive PubMed search was conducted of articles published from January 1, 2008, to June 26, 2020 in order to retrieve all available studies on this topic. Other relevant articles were identified from recent reviews and the bibliographies of the retrieved full-text articles.

Results

Among the 441 studies retrieved, 26 pediatric studies, comprising 15 case reports and 11 case series, used next-generation sequencing as a diagnostic tool. In these 26 studies, next-generation sequencing was performed on cerebrospinal fluid samples from 529 pediatric patients, and potential causal pathogens were identified in 22.1% of the cases.

Conclusion

There is increasing evidence that next-generation sequencing can play a role in identifying the causes of unexplained encephalitis, meningoencephalitis, and meningitis in pediatric patients, although the diagnostic value of next-generation sequencing is difficult to quantify. There is an increasing need for close collaboration between laboratory scientists and clinicians. We believe that further clinical studies should be performed to evaluate the performance of next-generation sequencing for individual targets and in high-risk populations.

Introduction

Central nervous system (CNS) infections caused by various microorganisms, including a wide variety of bacterial, parasitic, fungal, and viral pathogens, cause substantial morbidity and mortality, especially in pediatric patients.^1,2 In addition, it has been reported that at least a quarter of pediatric survivors have long-term neurological sequelae.³ The CNS is normally a sanctuary site because the blood-brain barrier offers protection from microorganisms; however, once an infection occurs in the CNS, it can spread to any site within it, eventually leading to inflammatory conditions such as meningitis, encephalitis, meningoencephalitis, and myelitis.^2,4 However, even in the best medically equipped developed countries, the specific pathogens of CNS infection cannot be identified by conventional methods in approximately 40% to 50% of overall cases and an even greater percentage of pediatric cases, which may lead to the inappropriate use of antibiotics or adverse outcomes.^1,2,5,6 Moreover, the low rates of etiologic identification impede the institution of evidence-based treatment protocols and the implementation of disease prevention strategies. Therefore, there is a critical need for timely, actionable diagnostic methods.

During the last 20 years, the development of molecular diagnostics as a complement to conventional methods such as serologic testing, immunohistochemistry, and culture has greatly improved our ability to identify pathogens. In particular, the California Encephalitis Project, which has been prospectively studying the clinical and epidemiologic characteristics of encephalitis since 1998, has made outstanding contributions in identifying its etiologies.^7,8 Among the diagnostic techniques employed, polymerase chain reaction (PCR) can only detect specific pathogens via specific probes and targeted primers. In contrast, unbiased next-generation sequencing (NGS) based on a high-throughput sequencing platform avoids this limitation, allowing for the comprehensive and quantitative analysis of all microorganisms present in a clinical sample.⁹ In recent years, NGS has been gradually recognized and applied to the diagnosis of CNS infections.^10,11 NGS involves the sequencing of total DNA or RNA in clinical samples followed by the analysis of millions of sequences, the removal of any reads mapping to the human genome, and comparison of the remaining sequences against known sequences to identify the potential pathogenic sequences (Fig 1). A critical requirement of this process is a robust bioinformatics platform to identify the presence of any pathogenic sequences. The use of cerebrospinal fluid (CSF) samples is a particularly attractive diagnostic option for patients with suspected CNS infection given the inherent risks and invasiveness of brain or meningeal biopsy. Published articles describing the effectiveness of NGS in identifying pathogens in pediatric patients with encephalitis, meningoencephalitis, or meningitis are limited to case reports and small retrospective or prospective case series.^5,12 The focus remains on the question of whether the diagnostic performance and yield of NGS testing for CNS infections supports the wider acceptance and adoption of NGS testing by clinicians.

In this narrative review, we summarized the existing evidence from pediatric patients, to evaluate the contributions and deficiencies of the NGS of CSF for the etiologic identification of unexplained CNS infections.