Metagenomic sequencing in encephalitis diagnostics: Challenges and opportunities in clinical settings

The primary aim of this study was to determine whether metagenomic next-generation sequencing (mNGS) can identify potential microbial agents responsible for encephalitis of unknown origin in immunocompetent patients, thereby enhancing clinical diagnostics. Cerebrospinal fluid samples from well-characterized patients (n = 17) diagnosed with encephalitis of unknown origin, according to Swedish national guidelines, were sequenced using mNGS using the Ion Torrent platform and analyzed using bioinformatic platforms. Samples from patients with known viral CNS infections i.e. HSV-2 meningitis (n = 4), VZV CNS infections (n = 3), enterovirus meningitis (n = 2), JCV CNS infection (n = 2) were used as controls for the methodology (n = 11). No viral agents were detected in 16/17 CSF samples from patients with encephalitis of unknown etiology. 13/17 CSF samples were analysed for the most common autoimmune antibodies and were negative. In one CSF sample from patients with encephalitis of unknown origin a Human pegivirus (HPgV) was detected. In 9/11 control CSF samples from patients with CNS infections, RNA or DNA of the known virus were detected. The main conclusion in this study was that the negative results were related to that the majority of included patients were immunocompetent. The finding of HPgV in a patient with unknown encephalitis was judged as a bystander. However, mNGS might detect more pathogens in other patient cohorts and this study implicates that a close collaboration between the clinical laboratory and the clinicians enables a safe implementation of metagenomics.