# Cambridge team tests AI-designed universal vaccine > Source: > Published: 2026-06-06 00:50:08.321049+00:00 # Cambridge team tests AI-designed universal vaccine According to the University of Cambridge and reporting in the Journal of Infection, pharmaphorum, and the BBC, a Cambridge-led team and spin-out DIOSynVax (DVX) Ltd have completed the first-in-human Phase 1 trial of pEVAC-PS, an AI-designed "universal" Sarbeco coronavirus vaccine. The University reports the open-label, dose-escalation study enrolled 39 healthy volunteers aged 18 to 50 and found the DNA-based, needle-free vaccine safe and well tolerated, with immune responses to SARS-CoV-2, SARS, and related bat coronaviruses. The antigen, a machine-learning-designed "super-antigen," was built from global viral surveillance data to target features conserved across the virus family. Reporting the paper, pharmaphorum notes the authors describe the immunogenicity as "modest" given participants' prior COVID-19 exposure, with no broad neutralising activity shown yet. Professor Jonathan Heeney said, "We've converted vaccine development from being reactive to being future proof." A larger Phase 2 trial is planned. ### What happened The University of Cambridge and its spin-out DIOSynVax (DVX) Ltd have reported the first-in-human Phase 1 trial of pEVAC-PS, described as the first vaccine whose active ingredient was designed entirely by computer simulation, according to the University of Cambridge and reporting in pharmaphorum and the BBC. The open-label, dose-escalation study enrolled 39 healthy volunteers aged 18 to 50 at National Institute for Health and Care Research (NIHR) Clinical Research Facilities in Southampton and Cambridge, sponsored by University Hospital Southampton NHS Foundation Trust, the University reports. The results are published in the Journal of Infection, and the work was primarily funded by Innovate UK. ### How the vaccine was designed Per the University of Cambridge, the team used machine learning on the global genetic sequence data for Sarbeco coronaviruses, logged by surveillance programmes worldwide, to design a synthetic "super-antigen" carrying features common across the whole virus group, including strains that have not yet emerged. The antigen was delivered as a DNA vaccine through a needle-free micro-jet system (the PharmaJet Tropis device, per the trial paper), which the University notes is compatible with most delivery platforms and could simplify mass vaccination. ### What the trial showed The University reports the vaccine was safe and well tolerated, and that volunteers developed immune responses to SARS-CoV-2, SARS, and related bat coronaviruses. Reporting the Journal of Infection results, pharmaphorum notes the immunogenicity was "modest but variable," which the authors link to participants' substantial pre-existing COVID-19 immunity, and that the candidate has not yet shown broad or robust neutralising activity. The authors say the measurable, cross-reactive binding to conserved sarbecovirus epitopes nonetheless supports the antigen-design concept. "We've converted vaccine development from being reactive to being future proof," said Professor Jonathan Heeney of Cambridge's Lab of Viral Zoonotics, the scientific lead. ### Why it matters Editorial analysis: For machine-learning and computational-biology practitioners, the trial is notable less for its immune readout than for validating an end-to-end pipeline that moves an algorithmically designed antigen from in-silico design through preclinical work to a first human safety study. It is an industry-level signal that computational antigen design can yield candidates testable in people; it is not a claim about the future roadmap of Cambridge or DIOSynVax beyond what has been reported. ### What to watch Editorial analysis: The decisive evidence will come from the planned larger Phase 2 trial, which the University says will test immune responses in a wider, more diverse population. Worth tracking: standardized neutralising-antibody titres, breadth against genetically diverse Sarbeco panels, durability, and transparency on the antigen-design algorithms, training data, and validation benchmarks that would determine whether the approach generalizes to other virus families such as influenza or the haemorrhagic fevers that DIOSynVax says are in its pipeline. ## Scoring Rationale A genuine world-first - the first vaccine with an antigen designed entirely by AI and computer simulation to reach a human trial - drawing broad mainstream and trade coverage, which makes it notable for applied ML and computational-biology practitioners. However, it is an early Phase 1 safety study in 39 volunteers with immunogenicity the authors describe as modest and no broad neutralising activity yet, leaving efficacy to a planned Phase 2. Scored in the upper notable band rather than major to reflect strong novelty and reach balanced against early-stage, modest results. Practice with real Health & Insurance data 90 SQL & Python problems · 15 industry datasets 250 free problems · No credit card [See all Health & Insurance problems](/problems/datasets/health)