Research work · 2022

Minimizing treatment-induced emergence of antibiotic resistance in bacterial infections

Science (New York, N.Y.)

Abstract

Treatment of bacterial infections currently focuses on choosing an antibiotic that matches a pathogen’s susceptibility, with less attention paid to the risk that even susceptibility-matched treatments can fail as a result of resistance emerging in response to treatment. Combining whole-genome sequencing of 1113 pre- and posttreatment bacterial isolates with machine-learning analysis of 140,349 urinary tract infections and 7365 wound infections, we found that treatment-induced emergence of resistance could be predicted and minimized at the individual-patient level. Emergence of resistance was common and driven not by de novo resistance evolution but by rapid reinfection with a different strain resistant to the prescribed antibiotic. As most infections are seeded from a patient’s own microbiota, these resistance-gaining recurrences can be predicted using the patient’s past infection history and minimized by machine learning–personalized antibiotic recommendations, offering a means to reduce the emergence and spread of resistant pathogens. Description Personal histories of past resistance A serious infection may initially be diagnosed as antibiotic susceptible but subsequently become drug resistant—and life threatening. Rather than de novo resistance mutation occurring, it is more likely that a resistant strain or species persisting in the patient’s gut or skin replaced the susceptible strain. From this starting point, Stracy et al. built machine-learning models that predict individual risks of gaining resistance to specific antibiotics using 8 years of records on more than 200,000 patients’ microbiome profiles (see the Perspective by Lugagne and Dunlop). Data on antibiotic use for urinary tract and wound infections were used to train the algorithms and to develop personalized antibiotic treatment strategies. For most patients, there was an alternative susceptibility-matched antibiotic that had a lower predicted risk of resistance emerging compared with the antibiotic prescribed by the physician. —CA Machine learning can help to identify personalized alternative antibiotic treatments.

Download PDF

DOI: 10.1126/science.abg9868