Almost a decade ago, in our report ‘Pathogen Genomics Into Practice’, we predicted that advances in whole genome sequencing would bring microbial genomic testing out of the laboratory and closer to the point of patient care.
Since then, a global pandemic – and fear of future ones – has catalysed the already growing interest in diagnostics for infectious diseases. One promising area is the use of rapid point of care testing to speed up diagnosis – an essential step to limiting the spread of an infectious disease.
Researchers at Guy’s and St Thomas’ NHS Foundation Trust (GSTT) have recently completed a study exploring the use of Oxford Nanopore Technologies’ (ONT) MinION, a portable benchtop DNA sequencing platform, for a rapid metagenomics test to diagnose lower respiratory tract infections, which can often lead to additional complications, including sepsis.
What is the problem?
Globally, infections of the lower respiratory tract (RTI) are the leading cause of sepsis. In 2017, this caused an estimated 1.8 million deaths that might have been prevented if the underlying RTI had been diagnosed and treated more quickly.
Respiratory tract infections can be caused by different types of agents which may be bacterial, viral or fungal. Effective treatment will differ according to the type of agent responsible. However, routine tests to identify the specific agent currently take 1-5 days depending on bacterial load and may only provide partial information.
This delay often leaves clinicians needing to make decisions about treatment before they know the cause of disease. Compounding this problem is the growing burden of antimicrobial resistance to existing drug therapies, which may further limit treatment options for sepsis infections.
This presents a challenge: how can clinicians be helped to make better decisions about treatment to prevent disease progression and sepsis?
Researchers at GSTT asked what is the feasibility and utility of introducing a routine metagenomics service to provide same day results to patients in intensive care with suspected respiratory infection.
What is metagenomic sequencing?
Metagenomics can be used to identify genes with known antimicrobial resistance (AMR) associations potentially supporting better prescribing, particularly of antibiotics. Metagenomics sequences the many microorganisms in a sample and uses bioinformatics methods to identify pathogens and the presence of drug resistance genes in the samples.
The GSTT team developed a 6-hour respiratory metagenomics (RMg) workflow using ONT’s long-read sequencing device. While there are several sequencing technologies available, the ONT devices are unique in their speed and therefore the potential of a rapid turnaround of results.
The test was then performed in a daily pilot service for patients with lower respiratory infections to assess its utility and feasibility of clinical implementation.
In the study, RMg provided same day results for 110 patients with a sensitivity of 93% and specificity of 81% when identifying pathogens compared to routine testing. This is significantly faster – hours instead of days – than routine antimicrobial susceptibility testing and suggests good concordance of findings.
Respiratory metagenomics contributed to prescribing in 88 of 110 cases (80%). In 24 cases (22%), antibiotics were started or escalated based on the presence of AMR associated genes. In 29 cases (26%), RMg informed de-escalation of treatment alongside routine testing. And finally, in 35 cases, while there was no change in treatment, RMg provided supportive evidence on prescribing decisions and reassurance of no unexpected pathogens.
This study did identify some limitations. As this was a pilot, they only had low sample throughput due to limited operator time. The relatively high sample failure rate (12%) was mostly due to issues in DNA extraction or operator-introduced contamination that could be prevented by using automation.
In addition, while there are databases describing known important AMR genes, these gene-phenotype associations are a work in progress and some important genes or variants resulting in AMR would be missed if we relied solely on RMg in clinical practice.
Opportunity and the future
The pilot demonstrates the feasibility of nanopore sequencing in ICU wards i.e. at the point of care. Albeit in a small cohort, evidence from this study suggests the potential to be reliable and actionable, making a difference in treatment decision-making. GSTT have now received additional funding to expand their programme to multiple hospital sites across the UK.
As the evidence builds for RMg, the accuracy and specificity of rapid metagenomics tests is likely to be limited by our knowledge of these gene- or variant-phenotypes, particularly for predicting antimicrobial resistance. Clinical use is likely to be in addition to standard AMR testing and this raises questions around the cost-effectiveness balanced against opportunities for time savings and better treatment decisions.