Globally, 5 million deaths are linked to antibiotic resistance every year. The World Health Organization (WHO) has identified antibiotic resistance as one of the most crucial challenges in modern medicine. We need to develop new and better ways to prevent bacterial strains from becoming resistant and find or create new antibiotics to treat the resistant strains that are threatening universal health. Artificial Intelligence(AI) is playing a significant role in advancing our ability to do this.
Researchers at such prestigious facilities as ADA Forsyth Institute, Stanford Medicine and McMaster University, Perelman School of Medicine at the University of Pennsylvania, NIH and MIT are using AI-powered learning algorithms to analyze vast datasets of bacterial genomes(the complete set of genes or genetic material present in a cell or organism). The speed at which AI is capable of analyzing and learning from this data is helping researchers to understand the wide variety of ways that bacteria can become resistant. This helps to predict and prevent antibiotic resistance, design new drugs to target specific bacteria, and predict how the body will respond to new antibiotics.
AI technologies are increasingly being used to improve the delivery and efficacy of antibiotics. These advanced systems can optimize dosing regimens, improve drug targeting, and monitor patient responses in real time. Examples of these include Vancomycin, a key antibiotic for the treatment of serious infections caused by Gram-positive bacteria, including methicillin-resistant Staphylococcus aureus (MRSA) and Amikacin, an aminoglycoside antibiotic, which is commonly used to treat severe Gram-negative infections. Both of these antibiotics require precise dosing to avoid ototoxicity(damage to the inner ear leading to hearing loss, balance issues,) and nephrotoxicity(damage to the kidneys). Treatment protocols created using AI learning strategies has shown high accuracy and specificity in predicting initial and subsequent doses of these drugs. AI systems have also been developed to monitor blood levels of these medications in real time and adjust dosing accordingly. They also ensure that optimal drug concentrations are maintained, thereby improving treatment efficacy and safety.
Colistin is an antibiotic of last resort for multidrug-resistant Gram-negative bacterial infections. Its use is limited because of the risk of significant nephrotoxicity. To combat this, researchers have used AI to develop targeted colistin delivery systems, such as nanoparticle-based delivery vehicles, which can transport the medication directly to the site of infection, to maximize therapeutic effects and minimize systemic toxicity. In addition, silver nanoparticles linked or paired to colistin have shown enhanced antimicrobial activity and reduced toxicity compared to colistin alone.
As I reported late last year, ADA Forsyth researchers have entered human clinical trials for the narrow-spectrum antibiotic FP100 that they developed in collaboration with Flightpath Bioscience to successfully eradicate the bacteria Fusobacterium nucleatum, which triggers gum disease, without harming the beneficial microbiomes in the oral cavity and stomach. Since Fusobacterium can travel throughout the body and cause disease, this has the potential to prevent systemic disease as well.
Sharing data, resources, and knowledge across institutions and countries is helping to improve the effectiveness of AI-based research in fighting antibiotic resistance. In July researchers at the ADA Forsyth Institute perfectly illustrated this when they parodied the movie Fight Club by stating “The first rule of Bacteria Fight Club: Always talk about Bacteria Fight Club.” In this instance, they wanted to share their research findings about the multidrug-resistant pathogen, Klebsiella, with the goal to develop better preventative measures and treatment protocols in hospitals throughout the world.
To find out why Klebsiella, a pathogen that exists in minute quantities in the nasal and oral cavity, was capable of causing so many hospital-acquired, drug-resistant infections, the researchers set up a “Bacterial Fight Club” by placing 200 species of oral bacteria in a glass jar and letting them starve to see which bacteria would survive. Klebsiella was consistently one of the winners because of its ability to take the nutrients it needed from the dead and dying bacteria that surrounded it. They also found that Klebsiella survive better in the nasal microbiome than in the oral microbiome, because of the decreased micro diversity of the nasal cavity. They brought to light the big role this seemingly minor bacteria plays in drug resistant hospital-acquired infections. Early detection and elimination of Klebsiella is critical, especially for people on ventilators, which provide ideal conditions for Klebsiella growth.
As we recognize the necessity of global AI research, to meet the challenge of Antibiotic Resistance and other global health issues, it is crucial that this research follows international standards for data collection and analysis.
Dr. Stephen Petras
An Illinois Licensed General Dentist