Small Phase 1 Safety Trial Tests First AI-Designed Coronavirus Vaccine Candidate

British researchers completed the first human safety trial of a vaccine whose active component was designed entirely by artificial intelligence. The trial ran from December 2021 to September 2023 and enrolled 39 volunteers who each received four doses. The vaccine uses a computer-generated protein, called a super-antigen, that mimics features shared by multiple Sarbeco coronaviruses.

Researchers compiled all publicly available genetic sequence data for these bat-borne viruses and applied machine learning to produce the protein. Administration does not require a needle. A microfluidic jet propels the antigen into the skin as a high-speed liquid stream, a method intended to reduce sharps waste and cold-chain requirements.

No significant safety concerns were reported across the four doses. Professor Saul Faust of the University of Southampton, the trial’s chief investigator, said viruses such as influenza, coronaviruses and Ebola evolve continuously and that current reactive systems struggle to keep vaccines matched to circulating strains.

Professor Jonathan Heeney of the University of Cambridge’s Lab of Viral Zoonotics, the scientific lead, said the approach converts vaccine development from reactive to future-proof and could protect against viruses that have not yet emerged.

Professor Marian Knight, Scientific Director at the UK’s National Institute for Health and Care Research, described the results as marking a pivotal leap in delivering broad, lasting viral protection. Researchers will now begin a phase 2 clinical trial to measure immune responses in a larger and more diverse population.