The needle-free DNA vaccine uses an AI-designed “super-antigen” to target shared features across the Sarbeco coronavirus family.
AI is moving deeper into healthcare, from patient-facing tools to drug discovery. Now, a University of Cambridge-led team is offering a look at what that could mean for vaccine design.
Cambridge researchers have reported early human safety results for a new vaccine designed with AI and computer simulations.
The Phase I trial tested a universal Sarbeco coronavirus vaccine candidate called pEVAC-PS, developed by Cambridge researchers and the university spin-out DIOSynVax. The findings were published in the Journal of Infection.
Sarbeco coronaviruses are a large group of related viruses. They include SARS-CoV-2, the virus behind the COVID-19 pandemic, as well as SARS and related viruses found in bats that could potentially spill over into humans.
The project aims to support broader protection against multiple members of this virus family, including future variants or related viruses that have not yet emerged in people.
The study involved 39 healthy volunteers between 18 and 50 years old. It was designed mainly to assess safety and tolerability.
The vaccine was well tolerated across all four tested dose levels, with no serious adverse events, serious adverse reactions or suspected unexpected serious adverse reactions reported.
The trial took place at the National Institute for Health and Care Research clinical research facilities in Cambridge and Southampton.
Unlike many traditional shots, this DNA vaccine was administered using a needle-free microfluidic jet system. This delivery method could make vaccination easier in some settings and may offer a more comfortable option for people who are uneasy about needles.
Needle-free vaccine delivery is also drawing interest in recent times. In 2025, Australian company Vaxxas raised about A$90 million to expand its work on vaccine skin patches. These patches use thousands of microscopic projections to deliver vaccine doses through the skin. The technology could offer a simpler, needle-free alternative to traditional injections.
What makes the Cambridge vaccine notable is how it was designed. Instead of building a vaccine around one known viral strain, the Cambridge-led team used AI and computer simulations to create a “super-antigen.”
Get industry leading pharma and biotech news, events and expert insights delivered to your inbox.
An antigen is the part of a vaccine that teaches the immune system how to recognize a threat. In this case, the AI-designed component targets features shared across the Sarbeco coronavirus family. The vaccine is designed to help the immune system recognize related viruses, even as they mutate.
The vaccine generated measurable immune responses to shared features across the Sarbeco coronavirus group, supporting the feasibility of the design. However, the overall immune response was modest, and the trial did not yet show the kind of broad neutralizing activity that would be needed to suggest wider protection.
Because participants already had some COVID-19 immunity, researchers could not fully separate the vaccine’s effect from immune responses shaped by earlier vaccination or possible virus exposure.
The next phase of research is expected to involve around 200 participants. This Phase II trial will include a wider and more diverse population and help researchers assess whether the AI-designed antigen can generate more consistent immune responses.
DIOSynVax is also applying the technology to other major health threats. Its pipeline includes work on seasonal flu, pandemic influenza threats, H5N1 bird flu and viral haemorrhagic fevers, including Ebola-related viruses.
The Cambridge work fits into a wider shift in AI-assisted medicine. In late 2025, the FDA qualified AIM-NASH, its first AI-based drug development tool, to help experts assess liver biopsy images in clinical trials for metabolic dysfunction-associated steatohepatitis (MASH). Investment is also moving into AI-led drug discovery. In 2026, Isomorphic Labs, a Google DeepMind spin-out, raised $2.1 billion to scale its drug design platform with the use of advanced AI models that study how drug molecules interact.
The Cambridge vaccine development remains early, but it shows one way AI could be used in future vaccine design.
FAQs
What is a universal vaccine?
A universal vaccine is designed to protect against a broader group of related viruses or strains, rather than one specific version of a virus.
Does this vaccine protect people from future pandemics?
Not yet. The Phase I trial mainly tested safety. Larger studies are needed to understand whether the vaccine can provide broad protection.
Why did researchers use AI to design the vaccine?
AI helped researchers identify features shared across related viruses. The goal is to train the immune system to recognize parts of the virus family that are less likely to change.
If you want your company to be featured on Xtalks.com, please email [email protected].
Download: Top 15 Pharma & Biotech Companies to Work For in 2026
Join or login to leave a comment
JOIN LOGIN