News

Jason Kwong, Department of Infectious Diseases, Austin Health, MDU PHL, Peter Doherty Institute for Infection & Immunity

NHMRC Early Career Fellowship awarded in 2017 for four years. 

Project
The MetaFIND (Metagenomics For INfectious Diseases) project: developing clinical metagenomics to improve the healthcare of patients with infections

Context
Clinical metagenomics is an exciting new approach using modern genome sequencing technology to analyse all the DNA in a clinical sample such as blood or a biopsy tissue sample, and determine all the organisms potentially causing an infection in that sample. Using clinical metagenomics, we hope to improve the healthcare of patients with suspected infections using this diagnostic approach, by identifying and characterising the genomes of the organisms causing the infections with a single diagnostic test, without the need to culture or grow the organisms in the microbiology laboratory.

Case example: a 14 year old boy presented to hospital with meningitis resulting in multiple seizures. Despite more than 50 diagnostic tests, the cause of his meningitis remained unknown. Due to the severity of his infection, he was commenced empirically on several antibiotics as well as immunosuppressive medications due to concern about an autoimmune condition causing meningitis. As a last resort, using clinical metagenomics, the boy was found to have a bacterial infection that was not able to be cultured in the laboratory. He was subsequently commenced on the appropriate antibiotic treatment and made a complete recovery.

Infections are among the most common causes of morbidity and mortality in modern healthcare, with between 33-50% of hospitalised patients being prescribed anti-infective treatment at any time. Accurate diagnosis of not only the clinical syndrome (e.g. meningitis), but also the infecting organism (e.g. Neisseria meningitidis – “meningococcus”) are critical to enable targeted, effective and efficient treatment. Without a clear diagnosis, clinicians have to “guess” the most likely cause or causes of the infection and start treatment presumptively.

Such empiric treatments are often excessively “broad-spectrum” to cover all likely possible infections, resulting in unnecessary patient exposure to potentially toxic antibiotics and promoting development of “superbugs” – infections resistant to our commonly-used antibiotics. Healthcare costs can be significantly inflated with empiric treatment, with the cost of broad-spectrum treatment for a severe infection due to a possible virus, bacterium, fungus or parasite potentially exceeding $1000 per day for the anti-infective treatment alone.

Culture (reproducing growth in a laboratory) of the infecting organism has historically been the cornerstone of infection diagnosis in hospital and clinical microbiology laboratories, particularly for bacteria and fungi, though there are a number of situations where this is difficult, unsuccessful or not possible. In these instances, identifying the infection relies heavily on DNA-based tests, such as nucleic acid amplification tests (NAAT), which search for a short DNA segment unique to a single organism species. Detection of that DNA segment indicates presence of that organism. However, these methods can erroneously miss some critical infections, and a result for one organism does not provide information about the presence or absence of another infection.

Metagenomics is the study of all the genetic material (DNA) present in a sample, including both human and infecting organism(s). Whole-genome metagenomic sequencing of clinical specimens has shown promise in diagnosis of culture-negative infections. In contrast to searching for a single gene target, whole-genome metagenomics examines all the DNA in the sample for multiple DNA signatures that can identify the organism.

It also facilitates reconstruction of the genome of the infection present, containing hundreds to thousands of genes.

There are a number of advantages of whole-genome metagenomics over other DNA tests such as NAAT. Firstly, it offers the ability to search for multiple viral, bacterial, fungal and parasitic infections concurrently from a single sample, alleviating the need to perform separate tests (often >20) for each possible infecting organism. Secondly, it provides the capacity to diagnose novel infections where other diagnostic tests have not yet been established.

Third, by reconstructing the genome of the infecting organism, additional detailed information such as the species present, the strain-type of the infection, the likelihood of transmission from other patients, and the presence of antibiotic resistance genes can be gleaned to inform epidemiological surveillance for outbreaks, infection control measures to prevent spread of the infection, and targeted anti-infective treatment.

Currently, using non-metagenomic methods, obtaining this information is not possible without in vitro culture of the organism. Metagenomics also offers improved sensitivity, as organisms can be identified by multiple DNA signatures, in contrast to a single target sequence with NAAT for example, which may have been altered or fragmented in the pre-testing process.