Clinical performance of real-time nanopore metagenomic sequencing for rapid identification of bacterial pathogens in cerebrospinal fluid: a pilot study

This study aimed to evaluate the usefulness of amplicon-based real-time metagenomic sequencing applied to cerebrospinal fluid (CSF) for identifying the causative agents of bacterial meningitis. We conducted a 16S rRNA amplicon sequencing using a nanopore-based platform, alongside routine polymerase chain reaction (PCR) testing or bacterial culture, to compare its clinical performance in pathogen detection on CSF samples. Among 17 patients, nanopore-based sequencing, multiplex PCR, and bacterial culture detected potential bacterial pathogens in 47.1%, 0%, and 47.1% samples, respectively. Nanopore-based sequencing demonstrated a sensitivity of 50.0%, specificity of 55.6%, positive predictive value of 50.0%, negative predictive value of 55.6%, and overall accuracy of 47.1%, compared to the gold standard method for bacterial culture. In 44.4% (4/9) of culture-negative cases, nanopore-based sequencing detected potentially causative pathogens, whereas four (23.5%) patients were positive only in culture. Using nanopore-based sequencing alongside bacterial culture increased the positivity rate from 47.1 to 70.6%. However, these values may be overestimated due to challenges in distinguishing significant pathogens from background noise. Meanwhile, the bioinformatics module in EPI2ME reduced the turn-around time to 10 min. Nanopore-based metagenomic sequencing is expected to serve as a complementary tool for pathogen detection in CSF samples by facilitating rapid and accurate diagnosis.