The rise in antimicrobial resistance (AMR) in bacteria is a threat to all people¹, especially those with weakened immune systems or undergoing surgery. Arguably one of the groups most at risk are preterm babies, with around one in 10 babies worldwide born premature² (a proportion that is rising). Because of their early arrival, these babies have organs and immune systems that are often underdeveloped. They also possess a disturbed gut microbiome, which further increases their vulnerability to bacterial infections. For those infants born with a very low birth weight (<1,000 grams), there is an added risk, as they tend to spend longer periods in neonatal intensive care units (NICUs) and receive more medical interventions. Up to 15% of these infants will suffer from a devastating gastrointestinal disease called necrotising enterocolitis (NEC), which is associated with an overgrowth of intestinal pathogens, leads to severe damage of the bowel wall, and is fatal in ~30-40% of cases. At the first signs of infection, clinicians will prescribe a first line antibiotic cocktail, while they await standard clinical microbiology results. These conventional diagnostic tests, based on isolation and cultivation of bacteria to determine species identification and antibiotic resistance profile, typically take at least 24 to 48 hours³. However, in the case of NEC, these timings are problematic as infants can go downhill very quickly and in many cases the bacterial ‘culprits’ are often resistant to numerous antibiotics.

To expedite testing of these at-risk babies, and to aid in potential diagnostics, an alternative approach would be to perform direct sequencing of stool samples, with real-time analysis of their microbiome via microbial/metagenomic and AMR profiling. This may allow clinicians to intervene in a timely fashion with antibiotics tailored to the bacterial infection. We explored this concept using the compact, real-time MinION Nanopore sequencer and a cross-disciplinary team comprised of scientists from the Quadram Institute, the Earlham Institute, the Natural History Museum, and clinicians from the Norfolk and Norwich University Hospital.

Development of a bespoke and freely available software, allowed us to reliably profile the preterm gut microbiota, and identify or ‘diagnose’ the potential causative bacterial species including their corresponding antibiotic resistance profile in less than 5 hours. Our methodology and computational tools were validated using known bacterial communities, and isolation, whole genome sequencing, and phenotypic antibiotic susceptibility testing of the causative pathogen Klebsiella pneumoniae, to benchmark this new workflow.  In the future, and with further large-scale clinical studies, data obtained from such analyses may allow clinicians to rapidly tailor specific antibiotic treatment strategies for at-risk patients, which could reduce drug side-effects, and is particularly important with the global threat of AMR. This study suggests that the MinION sequencer, combined with bespoke bioinformatic tools, may allow personalised health care strategies, empowering antibiotic stewardship for fragile premature infants, and perhaps further as the technique matures.

1          Organisation, W. H. World Antibiotic Awareness Week 2018. Think Twice. Seek Advice.  (2018).

2          Organisation, W. H. Preterm birth.  (2019).

3          Maugeri, G., Lychko, I., Sobral, R. & Roque, A. C. A. Identification and Antibiotic-Susceptibility Profiling of Infectious Bacterial Agents: A Review of Current and Future Trends. Biotechnol J 14, e1700750-e1700750, doi:10.1002/biot.201700750 (2019).