Kathryn Harris (United Kingdom)
Barts Health NHS Trust Barts Health NHS TrustAuthor Of 1 Presentation
UNTARGETED METAGENOMIC SEQUENCING FOR INFECTIOUS DISEASE DIAGNOSIS WITH A FOCUS ON ENCEPHALITIS
Abstract
Backgrounds:
Clinically, metagenomics have been most useful in the context of diseases that have poor diagnoses rates. Encephalitis is a prime example, where even in richly-resourced settings, the causative agent remains elusive in the majority of the cases. Here we present the results from 7 years of clinical metagenomics studies performed at the Great Ormond Street Hospital for Children, with samples obtained from in-patients and patients from around the UK and Europe.
Methods
Untargeted metagenomic DNA and RNA sequencing was performed for brain biopsies (fresh, frozen and FFPE) and CSF samples. Negative and positive sequencing controls were included in each sequencing run from 2017 onwards. Sequencing data preprocessing consisted of bioinformatic removal of low quality, low complexity and human sequences. Protein and nucleotide based similarity search was followed by probabilistic taxonomic classification with metaMix to infer the microorganisms present in the sample.
Results:
82 clinical samples from encephalitis patients were sequenced, consisting of 65 brain biopsies (11 FFPE, 54 fresh or frozen) and 17 CSF samples. Pathogens were detected in 24 samples, with no pathogens found in 48 samples. RNA was degraded in 10 samples. Immune system status was recorded for 49 patients (60% immunocompromised). In this subset of patients, we detected the causative pathogen in 13 cases, all immunocompromised patients. Finally, 18 tissue specimens from patients with other suspected infections were sequenced, resulting in similar rates of pathogen and no pathogen detection (39% each outcome, 12% RNA degraded).
Conclusions/Learning Points:
Metagenomics can play an important role in the management of hard to diagnose clinical syndromes, such as, but not limited to encephalitis. The effect is more striking in immunocompromised patients, as this is the patient population most at risk of infection with unexpected or novel microorganisms.