Clinical Evaluation of an Improved Metagenomic Next-Generation Sequencing Test for the Diagnosis of Bloodstream Infections. (1st June 2021)

• Record Type: Journal Article
• Authors: Jing, Chendi; Chen, Hongbin; Liang, Yong; Zhong, Ying; Wang, Qi; Li, Lifeng; Sun, Shijun; Guo, Yifan; Wang, Ruobing; Jiang, Zhi; Wang, Hui
• Is Part Of: Clinical chemistry. Volume 67:Number 8(2021)
• Page Start: 1133
• Page End: 1143
• Publication Date: 2021-06-01
• Subjects: Metagenomic next-generation sequencing -- bloodstream infection -- pathogen detection; Clinical chemistry -- Periodicals; Pharmaceutical chemistry -- Periodicals; Biochemistry -- Periodicals; Biochimie -- Périodiques; Diagnostics biologiques -- Périodiques; Biochemistry; Clinical chemistry; Pharmaceutical chemistry; Biochemistry; Laboratory Techniques and Procedures; Klinische chemie; Periodicals; 616.075605
• DOI: 10.1093/clinchem/hvab061
• Languages: English
• ISSNs: 0009-9147

Abstract

Abstract: Background: Metagenomic next-generation sequencing (mNGS) of plasma cell-free DNA has emerged as a promising diagnostic technology for bloodstream infections. However, a major limitation of current mNGS assays is the high rate of false-positive results due to contamination. Methods: We made novel use of 3 control groups—external negative controls under long-term surveillance, blood samples with a negative result in conventional tests, and a group of healthy people—that were combined and dedicated to distinguishing contaminants arising from specimen collection, sample processing, and human normal flora. We also proposed novel markers to filter out false-positive interspecies calls. This workflow was applied retrospectively to 209 clinical plasma samples from patients with suspected bloodstream infections. Every pathogen identified by the mNGS test was reviewed to assess the diagnostic performance of the workflow. Results: Our mNGS workflow showed clinical sensitivity of 87.1%, clinical specificity of 80.2%, positive predictive value of 77.9%, and negative predictive value of 88.6% compared with the composite reference standard. Notably, mNGS showed great improvement in clinical specificity compared with the current test while keeping clinical sensitivity at a high level. Conclusion: The mNGS workflow with multiple control groups dedicated to distinguishing nonpathogen microbes from real causal pathogens has reducing false-positive results. This contribution, with its optimization of workflow and careful use of controls, can help mNGS become a powerful tool for identifying the pathogens responsible for bloodstream infections.