A scientific research team from the University of Cambridge in the UK announced that for the first time, a new vaccine with core components designed entirely by artificial intelligence has been used in human trials, in an attempt to provide a broader prevention solution for possible future large-scale virus outbreaks and COVID-19-like epidemics.Researchers said that this is the first case where a key vaccine antigen is completely designed by artificial intelligence algorithms and entered into human clinical trials. The goal is not only to deal with the existing new coronavirus and its mutant strains, but also to include multiple coronaviruses that are currently mainly circulating among animals but have the potential to cross species to infect humans, so as to "get a head start" before the next pandemic occurs.
The concept of this vaccine candidate is different from traditional vaccines in that it does not target a specific strain, but attempts to build a line of defense against the entire coronavirus "family". The artificial intelligence system first received a large number of coronavirus genetic sequences collected by global surveillance projects. These sequences are equivalent to "instructions" for the virus, covering different types known to infect humans and animals. The algorithm then analyzes and pattern-recognizes this genetic information to design a so-called "super-antigen", which is used to train the body's immune system to produce an immune response with broad cross-reactivity capabilities. Even if the virus continues to mutate or the new coronavirus jumps from animals to humans, it can still recognize and launch defense to a certain extent.
In vaccinology, antigens are key components of vaccines and are what the body's immune system learns to "recognize and attack." Project leader Professor Jonathan Heeney of the University of Cambridge said that this is the first time that an antigen designed by artificial intelligence has been used in human trials. The performance of the related technology is "astonishing" and also shows "the huge potential of artificial intelligence in benefiting mankind." He said that what the team hopes to do is to develop vaccines that not only target "today's viruses" but also prevent future pathogens that may cause the next outbreak or pandemic, which will mean a "fundamental change" in the way we deal with pandemics.
Vaccines have played a key role during the COVID-19 pandemic, but the vast majority are based on the specific strains of the virus circulating at the time, with recipes updated as the virus continues to mutate. Many respiratory viruses, such as COVID-19 and seasonal influenza viruses, are good at changing their surface structure through mutation and thus evading immune recognition, which is why COVID-19 and influenza vaccines need to be updated regularly. "We are always one step behind," Heaney pointed out, and this attempt hopes to use new technologies such as artificial intelligence to identify "common weaknesses" in the virus family in advance and complete defense preparations before the virus mutates or spreads across species.
Currently, this AI-designed coronavirus vaccine candidate has completed the first phase of small-scale human trials, with a total of 39 subjects participating. The main purpose is to evaluate safety. Preliminary results published in the Journal of Infection showed the vaccine was described as "moderately" effective in stimulating an immune response, but still sparked a positive response within the industry. The research team has planned to conduct follow-up studies involving approximately 200 subjects to more comprehensively evaluate its immunogenicity and potential protection against different coronaviruses.
Saul Faust, a professor at the University of Southampton who participated in some of the clinical trials, commented that this technology of designing antigens based on artificial intelligence "really has potential" and described this research direction as "very exciting". He pointed out that when dealing with the constant mutation of viruses, especially potential pandemic pathogens, such technical methods are significantly better than traditional methods in the vaccine design stage.
The Cambridge team has not limited its target to coronaviruses. They are already conducting research on a universal seasonal flu vaccine in animal models, hoping to move away from the practice of updating flu vaccine formulas every year in the future. In addition, the team is also developing a vaccine for H5N1 avian influenza to respond to the situation if this virus, which is currently seriously harming poultry populations and is feared to have a pandemic risk, spreads widely among humans.
Researchers are also working on developing vaccine candidates for viral hemorrhagic fevers, including different types of Ebola virus. The current outbreak of hemorrhagic fever in the Democratic Republic of the Congo, caused by a species of Ebola virus for which there is no existing vaccine, creates a real urgency to develop a broader spectrum hemorrhagic fever vaccine.
Andy Pollard, director of the Oxford University Vaccine Group, who was not involved in the study, said this route had produced "very compelling evidence" in animal experiments. He called the data "quite fascinating" and said many scientists had not necessarily expected to be able to induce such a type of immune response in this way. Pollard also pointed out that the real test lies in the results of human trials, because the human immune system is far more complex than that of laboratory mice, shaped by years of natural infection, so the clinical effect remains to be seen.
From a broader perspective, many experts believe that artificial intelligence is expected to be a "game changer" in vaccine research. Artificial intelligence tools can not only be involved in designing antigens, but may also be used to predict the response of the human immune system to different vaccine candidates, thereby significantly speeding up the screening and development process, shortening the time from the emergence of the virus to the arrival of the vaccine, and "saving lives" in future public health events.
Marian Knight, scientific director of the British National Institute of Health and Care Research (NIHR), commented that the success of the early human trial of this AI-designed "superantigen" is a "transformational step" in achieving broad-spectrum, long-lasting virus protection. Lord Vallance, the British Minister of Science and Technology, called this development "another British scientific success story" and believed that it demonstrated how to transform scientific research advantages into new medical treatments with the support of artificial intelligence technology. He emphasized that as the first human trial results release positive signals, this work is expected to accelerate the pace of vaccine deployment in the long term and benefit more people around the world.