Rapid bacterial vibration test to fight AMR and sepsis

Rapid bacterial vibration test to fight AMR and sepsis

Harnessing bacterial nanomotions, this method swiftly tests antibiotic susceptibility without growth dependency, aiding precision treatment and combatting AMR globally.

A Swiss startup’s fast diagnostic method identifies effective antibiotics, potentially revolutionising sepsis treatment and combating antimicrobial resistance.

Antimicrobial resistance and sepsis

Antimicrobial resistance (AMR) is a critical health challenge today. Known as “superbugs,” drug-resistant bacteria cause millions of deaths annually. AMR occurs when bacteria evolve due to the misuse and overuse of antibiotics in medicine and animal husbandry. This makes infections harder to treat, increasing the risk of disease spread, severe illness, and death. Once easily treatable infections are now becoming resistant, necessitating the development of new antibiotics, which are slacking for many reasons. However, using these new antibiotics will foster AMR further, creating a troubling cycle. Thus, rethinking antibiotic administration in healthcare is essential to avoid exhausting treatment options.

The situation is especially critical in sepsis, a life-threatening condition where the body’s response to infection can rapidly lead to septic shock and death if not promptly treated. Combating bacteria-induced sepsis requires administering the correct antibiotic quickly. Traditional methods to identify an effective antibiotic can take several days. As a result, doctors often use broad-spectrum antibiotics, which target a wide range of bacteria but also contribute to the rise of antimicrobial resistance (AMR). Faster, more precise diagnostic tools are essential to ensure the correct antibiotic is prescribed.

How can we improve the treatment of sepsis patients in times of antimicrobial resistance? The rising cases of antimicrobial resistance make it difficult to determine which antibiotics are effective for patient treatment. Improved and faster diagnostics based on novel technologies that detect bacterial nanomotions can help identify the right antibiotic on time.

Alexander Sturm

Fast diagnostics based on bacterial vibrations

A Swiss startup has developed a technology that significantly reduces the time required to identify the appropriate antibiotic. This method measures nanoscale vibrations from bacteria due to their metabolic activity. When exposed to antibiotics, these vibrations change, indicating whether bacteria are susceptible or resistant.

An amplifier that combines the vibrations of a few hundred bacteria is used for measurements. This involves a cantilever, a “miniature diving board” supported on one end and free on the other, which deflects according to bacterial nanomotions. Unlike traditional methods, this nanomotion technology does not depend on bacterial growth, which is typically the time-consuming factor in antibiotic susceptibility testing. It currently provides results in as little as two hours.

Credit. Author

The role of machine learning

Measuring bacterial vibrations is a novel approach that generates vast amounts of data, which initially appear as noise. Machine learning algorithms are essential for interpreting these nanomotion signals. These algorithms analyse the vibration data and extract features indicating whether bacteria are resistant or susceptible to antibiotics.

Given the volume of data, machine learning is crucial. Classification models are trained on thousands of nanomotion recordings from various clinical isolates and antibiotics. These models are then tested on different bacterial strains to assess their ability to detect antibiotic susceptibility accurately.

The new technology can already test antibiotic susceptibility for two common sepsis pathogens, Escherichia coli and Klebsiella pneumoniae, with an accuracy of 89.5% to 98.9%, depending on the antibiotic. This method is significantly faster than standard hospital diagnostic tests.

As training data sets grow, including bacterial isolates from diverse geographic regions, the system’s accuracy and general applicability will improve. Initial clinical evaluation studies in a Swiss hospital are complete, with the publication of results expected by the end of 2024. Meanwhile, efforts are underway to expand the range of antibiotics and bacterial species tested within the two-hour classification model.

Impact on patient treatment

Bacterial nanomotions enable rapid antibiotic testing, potentially revolutionising the treatment of sepsis and other bacterial infections. By quickly identifying the most effective antibiotic, doctors can provide targeted treatment sooner, improving survival rates for sepsis patients. This approach also reduces the misuse of broad-spectrum antibiotics, combating antimicrobial resistance (AMR). For healthcare systems, this means more efficient use of resources and shorter hospital stays. Ultimately, this technology marks a significant advancement in the fight against superbugs.

Broader implications of the nanomotion technology

Nanomotion technology shows promise beyond sepsis, offering the potential for diagnosing other bacterial infections, such as those in the urinary tract, wounds, and lungs. This makes it a valuable tool in the global fight against antimicrobial resistance (AMR).

Researchers are exploring its use in phage diagnostics, an alternative method for combating bacterial infections. Further advancements in machine learning will enhance the interpretation of nanomotion data, increasing the technology’s effectiveness.

Since nanomotion measures a universal cellular phenomenon, it could also assess susceptibility to antimicrobials for Mycobacterium tuberculosis or antifungals for Candida albicans, which affects immunocompromised individuals. Additionally, this technology may lead to faster drug tests also on human cells, supporting cancer treatments and immunotherapies.


Journal reference

Sturm, A., Jóźwiak, G., Verge, M. P., Munch, L., Cathomen, G., Vocat, A., … & Cichocka, D. (2024). Accurate and rapid antibiotic susceptibility testing using a machine learning-assisted nanomotion technology platform. Nature Communications15(1), 2037. https://doi.org/10.1038/s41467-024-46213-y

Dr. Alexander Sturm is the Chief Scientific Officer of Resistell AG, a Swiss startup developing a nanomotion technology platform for diagnosing antibiotic susceptibility. In this role, he leads an interdisciplinary research team collaborating with European hospitals. During his previous research at ETH Zurich, Columbia University NY, and the Broad Institute of MIT and Harvard, he gained extensive expertise in infectious diseases and microbiology, including tuberculosis and salmonellosis.