Detection of hidden antibiotic resistance through real-time genomics - NCM 2024

Abstract Real-time genomics through nanopore sequencing holds the promise of fast antibiotic-resistance prediction directly in the clinical setting. However, concerns about the accuracy of genomics-based resistance predictions persist, particularly when compared to traditional, clinically established diagnostic methods. Here, we leverage the case of a multi-drug resistant Klebsiella pneumoniae infection to demonstrate how real-time genomics can enhance the accuracy of antibiotic resistance profiling in complex infection scenarios. Our results show that unlike established diagnostics, nanopore sequencing data analysis can accurately detect low-abundance plasmid-mediated resistance, which often remains undetected by conventional methods. This capability has direct implications for clinical practice, where such ‘hidden’ resistance profiles can critically influence treatment decisions. Consequently, the rapid, in situ application of real-time genomics holds significant promise for improving clinical decision-making and patient outcomes. Biography Ela Sauerborn is a clinical microbiologist in training at the TUM University Hospital Munich and is currently in the first year of a PhD (concurrent with clinical training) in Lara Urban's Genomics for One Health group at the Helmholtz AI Institute, focusing on nanopore sequencing for plasmid-mediated carbapenem resistance detection in a clinical context. In addition to her medical background, Ela holds an MSc in infectious disease epidemiology with a focus on mathematical modelling and a BA in social sciences.