A machine learning model with data from patients with infections identified the factors that contribute to antibiotic resistance in reoccurring infections.
The researchers noted that a patient’s history of infections and antibiotic treatments can be used with patient demographic data to predict which candidate antibiotics would likely prevent a reoccurring infection.
In many cases, bacterial infections are seeded from bacteria in a patient’s microbiota. The treatment of infections like this usually means a variety of antibiotics. Serious infections are often evaluated for antibiotic susceptibility which may guide the use of particular drugs. However, while the treatment may clear the initial infection, it is thought that antibiotic use may pave the way for resistant strains to replace the previous susceptible ones.
These infections are then diagnosed as antibiotic-susceptible and can reoccur becoming life-threateningly drug-resistant.
The study used a large longitudinal dataset of more than 200,000 Urinary Tract Infections, wound infections and the associated patients’ microbiome profiles. The research team looked for incidences where initial antibiotic treatment was not effective.
Bacterial infections
The aim was to understand why some infections became resistant while others did not. To do this, they carried out genomic sequencing of patient bacteria in individuals who experienced early UTI reoccurrence providing a detailed view of the strains and species of the original and comparing them to the reoccurring bacteria.
The researchers discovered that the resistant infections were caused by strain replacement, not by point mutations in the original strain. Using the findings from the data, the team developed a machine learning model that can predict the risks of a pathogen gaining resistance to particular antibiotics at an individual patient level.
Jean-Baptiste Lagagne and Mary Dunlop in a related Perspective: “Machine learning recommendation systems such as the one presented [here] have the potential to substantially improve patient outcomes and could play a major role in mitigating antibiotic resistance,” write
They added: “This analysis reveals an underappreciated path to reinfection, with the original species being treated and eliminated but with the treatment ultimately setting the stage for other resistant strains to emerge.”
