Bacterial infections remain a major threat to public health, and doctors need better ways to spot and track infections in patients. Now, researchers have created a new imaging agent that can accurately visualize infections from a dangerous family of bacteria deep within the human body.
The new study was published in the April 14 issue of Science Translational Medicine. The imaging agent can be used with standard positron emission tomography (PET) techniques and detects infections from Enterobacterales, a group of bacteria that includes common pathogens such as Escherichia coli.
The agent was safe when tested with 26 patients and allowed the scientists to track how the patients responded to antibiotic treatment. It also differentiated bacterial infections from inflammation and tumors, and from lung inflammation due to COVID-19 in hamsters.
"Our method could complement current approaches used for the diagnosis and monitoring of infections, especially in high-risk patients where current methods may not work well, or where the risks of missing an infection are very high," said Sanjay Jain, professor of pediatrics at the Johns Hopkins University School of Medicine and senior author of the new study.
The new technology could also empower "precision medicine" approaches to best design and customize medical treatments for individual patients, he added.
A Persistent and Elusive Public Health Threat
Enterobacterales is the largest group of disease-causing bacteria in humans, encompassing pathogens such as E. coli, Salmonella, and Klebsiella pneumoniae. These microbes can cause meningitis, brain abscesses, pneumonia, and other life-threatening diseases.
Many species of Enterobacterales have become increasingly resistant to common antibiotics over the past few decades, according to the study. The rise of antibiotic resistance has led the Centers for Disease Control to label some drug-resistant Enterobacterales as urgent threats to human health.
A quick diagnosis is essential when treating bacterial infections, and doctors may have to monitor or change treatment if patients with drug-resistant strains don't respond to antibiotics. This depends on being able to visualize where the bacteria reside in the body, but infections with Enterobacterales pose unique challenges that complicate imaging and diagnosis.
For example, tests that study blood, urine, or stool often fail to detect Enterobacterales infections that appear in inaccessible sites deep within the body, Jain said. Other approaches such as biopsy or surgery are invasive and time-consuming, and can still miss the site of infection.
"Moreover, current noninvasive imaging methods such as CT or MRI provide non-specific readouts and cannot differentiate infections from non-infectious processes such as cancer or autoimmune diseases," he added.
Jain and Alvaro Ordonez, an assistant professor at the Johns Hopkins University School of Medicine and lead author of the new study, and colleagues tackled these difficulties with 18F-FDS, a radioactive molecule they had previously developed and tested in animals. These types of molecules are often used in medical imaging and are safe for patients.
The molecule takes advantage of a sugar that many bacteria rely on as an energy source, but that human cells don't use. After ingesting the sugar, the bacteria show up as bright spots during standard PET imaging.
The scientists previously found that Enterobacterales bacteria ingested the agent in rats, allowing the team to visualize entrenched infections. In the new study, they took the next step and tested18F-FDS in 26 patients, who harbored either infections with Enterobacterales or other medical conditions such as cancer and inflammatory diseases.
Tracking Down Infections in Patients and Animals
After injecting the agent into the patients, the researchers saw it safely and rapidly identified infections in the lungs, skull, and other sites during PET imaging. The tests detected both drug-susceptible and drug-resistant Enterobacterales, and the patients showed no adverse effects from the injections.
The agent also didn't highlight lung cancer lesions and sites of sterile inflammation, which can be mistakenly flagged as infections by current imaging methods.
To monitor treatment responses, the team repeated PET imaging in 13 of the patients after they completed antibiotic regimens. The images allowed the scientists to monitor how well the antibiotics killed the bacteria, and identified patients with drug-resistant strains that didn't respond to treatment.
Ordonez and colleagues then applied 18F-FDS to hamsters with either lung inflammation from SARS-CoV-2 or pneumonia caused by K. pneumoniae bacteria. Patients with bacterial pneumonia must often receive additional antibiotics, making it important to be able to distinguish it from nonbacterial inflammation, Jain said.
The method successfully highlighted the K. pneumoniae infections without showing up in the hamsters with nonbacterial lung inflammation, suggesting the platform could help clinicians pinpoint secondary bacterial infections in hospitalized patients with COVID-19 and other viral infections.
However, a new imaging agent must also be practical for use in the clinic. Many radioactive molecules used in PET imaging have short half-lives, meaning they must be synthesized either at or nearby the medical facility, according to Jain. This requirement limits the available supply, as synthesis requires expensive infrastructure and trained personnel.
With this in mind, the researchers also developed a cartridge system that allows scientists to rapidly synthesize 18F-FDS from a precursor molecule named 18F-FDG in under 10 minutes. This precursor is the most widely available imaging agent and is available at every PET imaging site worldwide, according to Jain.
"Therefore, the use of our one-step room temperature cartridge system will greatly increase the clinical adoption of this new radiopharmaceutical globally as well as allow on-demand supply of 18F-FDS," he said.
The team cautioned that their method has several limitations, as it failed to visualize a large liver abscess in a patient and generated signals in some fluids in the body that were probably uninfected.
Jain added that their study had a small sample size, and called for larger experiments with more patients to confirm the new findings. He said his team is gearing up to perform these studies alongside other collaborators around the world.