World’s first AI-designed vaccine is here: Could it stop future pandemics?

Artificial intelligence is increasingly being tested across healthcare for implication in disease detection as well as drug discovery.

Now, scientists say it has helped create something that could transform how the world prepares for future outbreaks: a vaccine designed not for a single virus, but for an entire family of viruses.

Researchers at the University of Cambridge, working alongside biotechnology company DIOSynVax, have developed what they describe as a new category of vaccine built using artificial intelligence and machine-learning systems.

The technology aims to generate protection not only against existing pathogens but also against future variants and even related viruses that have not yet infected humans.

More from Health

Unlimited curiosity, uncertain outcomes: The risks of AI-augmented mass inquiry

India’s digital sovereignty depends on balancing AI innovation and regulation

The development has already cleared an important milestone. A first-in-human clinical trial has demonstrated that the experimental coronavirus vaccine is safe, while larger studies involving more than 200 people are expected to follow.

Scientists involved in the project argue that the technology could eventually help the world move away from constantly reacting to outbreaks and instead prepare for them before they occur.

The same platform is now being explored for influenza, bird flu and Ebola-related diseases.

Is this a new chapter in vaccine development?

For decades, vaccines have largely been designed to target a particular virus strain that is already circulating. While this approach has saved countless lives, it faces a major challenge which is the fact that viruses constantly evolve.

As pathogens mutate, vaccines often require updating. The annual reformulation of seasonal influenza vaccines and the repeated modification of Covid-19 vaccines are prominent examples of this challenge.

Researchers behind the new technology believe the existing model leaves health systems perpetually trying to catch up with viruses.

Professor Jonathan Heeney, from the lab of viral zoonotics at the University of Cambridge's Department of Veterinary Medicine, said, "What that Covid pandemic taught us is how fast we can make vaccines, but we're still using the old paradigm.

"This is about making one vaccine that will get them all based on their relationships."

Scientists involved in the project describe the technology as a significant departure from conventional vaccine development because it attempts to identify common biological features shared across an entire virus family rather than focusing on a single strain.

Researchers believe such an approach could provide protection against multiple variants simultaneously and potentially offer defence against related pathogens that may emerge in the future.

How did artificial intelligence help create the vaccine?

Scientists gathered coronavirus genome sequences collected through surveillance programmes operating across different parts of the world.

These datasets included information from previous outbreaks, current circulating viruses and animal viruses that have attracted scientific attention because of their potential to infect humans.

Artificial intelligence systems then examined these genetic blueprints to determine which parts of the viruses remain relatively unchanged across the broader coronavirus family.

The objective was not simply to identify similarities. Researchers wanted to pinpoint components that are so important to viral survival that they cannot easily mutate without weakening the virus itself.

The result was the creation of what researchers call a "super-antigen". In vaccine science, antigens are the components that teach the immune system what to recognise and attack.

Rather than selecting an antigen from one known virus, the Cambridge team used AI-driven analysis to design an antigen intended to represent common characteristics found across many related coronaviruses.

Explaining the process, Heeney said, "You hoover up all the genomic sequences; what's known from around the world, from past outbreaks and current outbreaks, and you do some basic structural science.

"We take all these different sequences…and we think, 'OK, what's consistent amongst them, what's not changing, what is essential for their life' and that's what we target.

"It not only predicts, but it targets what is essential for that virus family. We're targeting something in a virus family, which the virus can't change easily."

The researchers believe this strategy could allow vaccines to remain effective even when viruses undergo significant genetic changes.

What is the universal Sarbeco vaccine?

The vaccine tested in humans is known as a universal Sarbeco coronavirus vaccine. Sarbecoviruses are a subgroup of coronaviruses that includes SARS-CoV-2, which caused the Covid-19 pandemic, as well as the virus responsible for the SARS outbreak in the early 2000s.

The group also includes several coronaviruses circulating in bat populations that scientists monitor because of their potential to jump into humans.

Instead of concentrating on one member of this viral group, the vaccine combines characteristics that are shared across the broader family.

Researchers say this allows the immune system to recognise a wider range of related viruses rather than a single target.

What happened during the first human trial?

Published in the June edition of the Journal of Infection and sponsored by the University Hospital Southampton NHS Foundation Trust, the Phase I trial was designed primarily to assess safety.

The study involved healthy volunteers between the ages of 18 and 50. Reports indicate that between 39 and 49 participants were enrolled across research sites in Cambridge and Southampton.

Participants received the vaccine as a DNA-based formulation delivered through a microfluid jet system. Unlike traditional injections, this method does not use a needle. Instead, a highly pressurised stream of liquid passes through the skin and delivers vaccine material directly into cells.

Researchers reported that the vaccine was safe and generated immune responses against multiple coronavirus targets.

Importantly, immune activity was observed not only against SARS-CoV-2 and SARS but also against related bat coronaviruses that have not infected humans on a large scale but are considered potential future threats.

The Journal of Infection findings described the immune response as "modest". However, scientists pointed out that the purpose of a Phase I trial is to establish safety and gather preliminary information rather than provide definitive proof of effectiveness.

The next phase of testing is expected to recruit more than 200 participants, allowing researchers to better evaluate how the vaccine shapes immune responses.

Why do researchers believe this could help prevent future pandemics?

By the time scientists identify a new pathogen, design a vaccine, complete testing and distribute doses, the virus may already have spread extensively.

Researchers believe universal vaccines could narrow that gap.

Heeney said, "There's a lot of viruses out there, and once we know them, we start chasing them, but we have to change that paradigm.

"And that's what this is about, it's about making vaccines that not just protect us from today's viruses, but the ones that haven't yet happened."

The project is built around the idea that surveillance programmes already collect vast amounts of information about viruses circulating in animals and humans.

The ultimate goal is to create vaccines in advance of outbreaks rather than during them.

Could the same tech work against Ebola and bird flu?

Researchers are already applying the platform to several other viral threats. Among the most significant is bird flu, particularly H5N1, which has spread widely among bird populations and has also infected mammals.

Heeney described the threat this way, "This is because this bird flu virus is all around the world, on most continents, not only impacting birds, but also mammals and humans, and it's even in the food chain in America, in milk.

"So it's quite a worry, but there's different clades, or types of families, and there's particular ones in south-east Asia that are very different, but that have killed people in double digits.

"So it's about making sure that our technology can get whatever is going to pop up and protect us, and to get ahead of that curve, instead of chasing it."

Scientists are also investigating whether the same approach can be used against viral haemorrhagic fevers, including Ebola viruses. The issue is particularly relevant because different members of the Ebola family can cause outbreaks. Existing vaccines do not necessarily cover every species.

Referring to current outbreaks in the Democratic Republic of the Congo and Uganda, Heeney said, "A great example of that is what's going on now in the DRC. Again, yet another Ebola virus, but it's not the same one, it's from the same family. Deja vu. We're behind the curve, and these viruses belong to the same family.

"So, what we're trying to do is to make a vaccine that will protect against all those different viruses in a family, and it's a big paradigm change."

He has also pointed to the devastating West African Ebola epidemic between 2013 and 2016 as an example of how difficult it can be to respond quickly once an outbreak is underway.

The researchers say the same technology could potentially offer protection against thousands of variants within viral families.

Also Watch:

With inputs from agencies