AI-Designed Vaccine Shows Promise in Human Trials

A vaccine developed using artificial intelligence has successfully completed its initial human clinical trial, with potential to offer broad protection against various coronaviruses and prepare for future outbreaks.

Researchers from the Universities of Cambridge and Southampton have engineered a so-called “universal vaccine,” which aims to guard against multiple sarbecocoronaviruses, including SARS-CoV-2, the virus responsible for the pandemic. The traditional vaccines require constant updates to accommodate evolving virus strains, described by Professor Saul Faust from the University of Southampton as a “dog chasing its tail.”

Faust pointed out that viruses like influenza and Ebola change so rapidly that by the time a vaccine is ready, it may not be effective. Current vaccines, termed ‘reactive,’ simply can’t keep pace with these mutations.

The research team used AI to compile all known genetic sequence data for sarbecocoronaviruses, creating “superantigens” — components that are common among these viruses. This approach potentially addresses those that haven’t even emerged yet.

In clinical trials, the vaccine was found to be safe and triggered an immune response in 39 healthy participants. This marks a significant milestone as it’s the first vaccine entirely designed via computer simulation to be tested on humans.

In a novel delivery method, the vaccine is administered through a microfluidic jet, enabling it to penetrate the skin without the need for needles. Researchers believe this could streamline mass vaccination efforts.

Faust remarked that this universal vaccine class is prepared for the future, suggesting it could simultaneously protect against multiple variants and even unidentified viruses. The hope is that advanced development of such vaccines could prevent viral outbreaks, save countless lives, and help maintain economic stability.

However, there are concerns about the role of AI in medicine, particularly regarding the potential for biased data. Some experts warn that AI can produce misleading results or “hallucinations,” complicating accountability in healthcare failures. Privacy and the necessity for human oversight in evaluating patient health history versus single data sets also remain critical discussions in the field.

Researchers acknowledged the importance of larger clinical trials involving diverse populations, emphasizing that their findings were published in the Journal of Infection.