 |
| Image created by Dr. Michael J. Miller |
For patients with bacterial infections, the sooner they are treated with the appropriate antibiotics, the better they will fare. Current methods for determining which drugs might work for each patient rely on growing bacteria from the patient in the lab and take days to yield results. In the meantime, patients are often given broad-spectrum antibiotics, which encourage drug-resistant infections, a significant public health threat.
An innovative diagnostic approach developed by researchers at the Broad Institute of MIT and Harvard and at Massachusetts General Hospital could one day help patients get the most effective treatment faster, which could reduce the need for broad-spectrum antibacterials.
The method, called Genotypic and Phenotypic Antibiotic Susceptibility Testing through RNA detection, or GoPhAST-R, analyzes the growth and genetic activity of the bacteria to quickly determine the pathogen's susceptibility to various medicines. Now, the researchers have demonstrated the approach's potential in a pilot study on blood cultures from patients undergoing inpatient treatment for infections.
The paper is published on the pre-print server medRxiv. The findings will appear in the Journal of Clinical Microbiology.
"With this study, we've taken another step toward our broader goal, which is to help people make diagnoses faster so that patients can get better faster," said study senior author Roby Bhattacharyya, an associate member at the Broad Institute of MIT and Harvard, an assistant professor at Harvard Medical School, and an attending physician in the Infectious Diseases Division of Massachusetts General Hospital (MGH) Department of Medicine.
In the GoPhAST-R method, researchers expose bacterial samples to various antibiotics and then use an RNA detection platform to look for distinct patterns of change in messenger RNA expression, which reflect differences in the activity of bacterial genes.
These mRNA changes appear mere minutes after antibiotic exposure in bacteria that are drug-susceptible, but don't arise in drug-resistant ones. In addition, the method examines genes that are known to underlie antibiotic resistance, which can give clues to the underlying bacterial mechanisms and point to potential therapeutic options.
In previous work on a small number of patient samples from MGH's clinical microbiology laboratory, the research team showed that GoPhAST-R can determine antibiotic susceptibility less than four hours after bacteria are detected in a blood culture, compared to 28-40 hours using standard clinical laboratory methods.
In the new study, they expanded the clinical pilot to include blood samples from 42 patients hospitalized with infections from Escherichia coli or Klebsiella pneumoniae, two of the most common pathogens seen in bloodstream infections.
The researchers exposed the blood cultures to nine different medicines from three different antibiotic classes and later performed transcriptional profiling on the NanoString platform. They examined mRNA changes in 10 genes for each antibiotic class, plus a handful of genes that confer resistance to beta-lactam drugs.
Their results achieved 95% agreement with those from the gold-standard growth-based assays for antibiotic susceptibility. "In this pilot, we've shown that GoPhAST-R is an approach that can work well in the clinic," Bhattacharyya said.
"The next step will be to show GoPhAST-R's utility in making decisions about patient care in real time. We'd love to one day see it as an assay that can be employed in hospitals everywhere, to help patients get more effective treatments without promoting drug resistance," he added.
Bhattacharyya acknowledged the participation of the MGH clinical microbiology lab in making the study possible, along with the contributions from patients in the hospital. "We absolutely couldn't make these discoveries without them."
More information: Eleanor L. Young et al, Clinical Pilot of Bacterial Transcriptional Profiling as a Combined Genotypic and Phenotypic Antimicrobial Susceptibility Test, medRxiv (2024). DOI: 10.1101/2024.07.10.24310021.