Latest News

Digital Antibiotic Susceptibility Testing: A More Efficient Way To Test Antibiotic Resistance

By Benjamin Ross

April 6, 2017 | Travis Schlappi sees the stewardship of antibiotics as a point of medicine that is drifting by the wayside. When a patient comes into a doctor’s office, it is often the case that the doctor is not able to precisely determine the antibiotic resistance of the infection within the timespan of the initial visit. The result is that the doctor will prescribe a cocktail of antibiotics, hoping that at least one of them will kill the bacteria. But this method of treatment runs the risk of increasing antibiotic resistance.

It’s a problem that Schlappi, a PhD candidate in Chemical Engineering at the California Institute of Technology, has focused a great amount of time and effort on fixing. The solution he and his team at Caltech developed, a method for determining antibiotic susceptibility within the time frame of a doctor’s visit, caught the attention of the judges of the inaugural Early Innovator Award. The panel honored Schlappi at the Molecular Medicine Tri-Conference (MMTC) in San Francisco, Calif. in February.

Schlappi and the team at Caltech’s solution is an extension of an article they published in Angewandte Chemie in 2016. The method, named “digital antibiotic susceptibility testing” (dAST), determines the phenotypic antibiotic resistance or susceptibility for a bacterial species of interest, comparing the growth rate of bacteria on antibiotics with that of bacteria not exposed to antibiotics. With this method, doctors can determine the antibiotic resistance profile of an infection before prescribing an antibiotic.

DXX-Schlappi“Our new idea was to measure DNA replication instead of cell growth, and measure DNA replication with a very precise method (digital PCR),” Schlappi wrote in his entry. “The precision of digital nucleic acid amplification enables the antibiotic exposure time to be reduced to only 15 min.”

The dAST method was used in a 30-minute sample-to-answer antibiotic susceptibility test (AST) on the E. coli from a clinical urinary tract infection (UTI) sample; 15 minutes was for sample collection, and 15 more for analysis. The work from this study, while currently under review for publication, was displayed at MMTC for the Early Innovator Award judges.

“Because the pathogen of interest for around 80% of urinary tract infections is E. coli, we chose E. coli as our target pathogen to show that the dAST method works,” Schlappi told Diagnostics World via email. “For other pathogens, the dAST method could be readily applied by developing nucleic acid primers for that pathogen of interest.” The dAST method as presently developed will tell you that there is E. coli or no E. coli in the urine sample, but it doesn’t tell you if there is a different infectious bacteria in there. Future iterations may include an identification step where any of the common bacterial species can be identified.”

The 30-minute process of digital nucleic acid amplification is much faster than the normal methods used by today’s doctors, and that’s putting it mildly. The current methods of measuring the response of bacteria to antibiotics, either by optical density (OD) or a colony plate, usually takes from 24-72 hours to yield intelligible results.

While the dAST does wonders in the realm of time-saving and efficiency, it is not, unfortunately, a miracle worker. The focus of the test is limited to bacteria with doubling times from 10-30 minutes. This limitation works effectively on species like E. coli, but when it comes to slow growing species like tuberculosis, which doubles every 3-4 days, the dAST proves inadequate.  

Schlappi and his colleagues hope their method has a chance of being commercialized and translated into clinical practice. Several things must happen before that hope can become a reality. Their next step is to integrate the solution into an autonomous device that an untrained user could operate.

Then Schlappi hopes to adjust this dAST method to other samples and antibiotics. “As I developed my solution, I found the variability in biological samples quite surprising,” he wrote in his entry. “Just because the species in question is E. coli, it doesn't mean they will all act the same. Our method relies on bacteria growing robustly and quickly when no antibiotic is present; dealing with clinical urine samples was definitely a surprise as different clinical samples had bacteria with different growth rates and different antibiotic effectiveness.”