According to MIT researchers, booster injections and seasonal variation doses may become obsolete thanks to a novel approach to vaccination that incorporates machine learning. By targeting infected cells, this “pan-variant” vaccination would quickly control infections while ignoring the virus itself.
To be clear, this is currently undergoing animal testing and has not yet been put into use. Therefore, longer-lasting remedies than sporadic boosters for particularly problematic strains are required as COVID becomes a resident virus in the human population.
The issue is that, despite how fantastic mRNA vaccines are, they are reactive rather than proactive: after spotting a variation, you sample its spike protein or another distinguishing property and introduce it to the immune system so it is alerted to look out for it. Like letting a search and rescue dog inspect a lost hiker’s belongings.
Finding a different, longer-lasting method to protect the body against COVID attack was the goal of researchers at MIT’s Computer Science and Artificial Intelligence Laboratory. Their research was presented in a study that was released today in the journal Frontiers in Immunology.
The scientists decided against trying to combat the virus itself because its most identifiable characteristic, the spike protein, is constantly evolving. They concentrated instead on specific chemical signals that consistently show up on the surface of virus-infected cells. The infection would be stopped before it progressed to a risky or even infectious level if these could be identified early and the immune system‘s T cells mobilized swiftly.
These surface markers, known as human leukocyte antigens, act somewhat like raising semaphore flags by presenting a range of peptides to T cells. If everything is in order, the T cell continues on and the typical mix of well-known peptides appears. If there is a problem, the T cells may start firing after a viral fragment is hoisted up the flagpole.
What role does machine learning play in all of this, then? There is a wealth of information available that lists the different proteins and amino acid chains found in COVID, what they transform into once inside a cell, and how infected cells use HLAs to signal this.
When sorting through large amounts of noisy data in search of certain combinations of attributes, optimization challenges like these are amenable to the use of machine learning methods. In this instance, they used algorithms to catalogue the pertinent peptides and chose roughly 30 that are present or “conserved” in all virus variants, as well as being linked to HLAs, and are most likely to be employed as flags for T cells to recognize.
Transgenic mice given our HLA variants and this novel vaccination displayed a significantly more robust immune response in the short term following infection, and none of them succumbed to the virus.
This work provides evidence in a living system, a real mouse, that the vaccinations we developed using machine learning can provide protection from the COVID virus, according to Brandon Carter, an MIT PhD student and one of the paper’s lead authors.
An intriguing potential advantage is that immunocompromised individuals might gain significantly from this strategy even though mRNA vaccinations are ineffective for them. The “Long COVID” victims may also find some relief from a more thorough immune attack on their particularly resistant virus.
As stated in the study’s abstract:
Specific T cell responses alone can successfully decrease the pathogenesis of SARS-CoV-2 infection, as shown by the absence of specific antibody response in MIT-T-COVID-immunized mice. Our findings imply that more research on pan-variant T cell vaccines is necessary, particularly research for people who are unable to make neutralizing antibodies or for Long COVID prevention.
It’s a fruitful area of research and an excellent method to use computer technology advancements for the benefit of world health. But, it’s also crucial to remember that the “pan-variant” option is still in its infancy. One possibility is that it could complement or compete with already available vaccines; what if, for example, mRNA priming targets the best peptides for the immune response vaccination and destroys them? They would be working against each other. Furthermore, an excessive immune response increases the chance of collateral harm, such as the incorrect removal of cells that are signaling in two different ways.
Nonetheless, these are the right questions to ask because it appears that the new vaccine’s fundamental purpose is being served. As the team conducts additional testing of this intriguing strategy, we will learn more.
