“/>Image Credits: Achiko
According to MIT researchers, a new approach to vaccines with a machine learning twist could put an end to boosters and seasonal variations. This “pan-variant” vaccine would ignore the virus itself, but quickly bring infections under control by targeting infected cells.
To be clear, this is still in animal testing and a long way from being deployed. But as COVID becomes a resident virus in the human population, there is a demand for more sustainable solutions than occasional boosters for particularly bad species.
The problem is that as amazing as the mRNA vaccines are, they are reactive, not proactive: you see a variant, you taste the spike protein or some other distinguishing feature, and you slip it into the immune system so it knows it’s it’s your turn. look after. It’s a bit like letting a rescue dog sniff a lost hiker’s belongings.
Researchers at MIT’s Computer Science and Artificial Intelligence Laboratory wanted to find another, more sustainable way to protect the body against COVID attacks. An article describing their findings was published today in the journal Frontiers in Immunology.
The team decided to hit on the idea of attacking the virus itself, because its most distinguishing feature, the spike protein, is always changing. Instead, they focused on certain molecular signals that appear reliably on the surface of cells infected by the virus. If these could be caught early and the immune system’s T cells deployed quickly, the infection could come to a halt before it reaches dangerous or possibly even infectious levels.
These surface signals, called human leukocyte antigens, present a variety of peptides to T cells, a kind of semaphore flags. If all is well, it’s the usual combination of known peptides and the T cell moves on. If something goes wrong, a fragment of the virus can be hoisted up the flagpole and the T cells open fire.
So what does machine learning have to do with this? There’s a lot of data cataloging the different proteins and amino acid chains in COVID, what they turn into once they enter a cell, and how the cells use HLAs to indicate they’re infected.
Machine learning algorithms are good at solving optimization problems like this one, which involves searching through a lot of noisy data for specific combinations of qualities. In this case, they designed algorithms to catalog the relevant peptides and selected about 30 that are present or “conserved” in all versions of the virus, but are also associated with HLAs, and are likely to be used as flags for T cells to see.
Transgenic mice given our versions of HLAs and this new vaccine showed a much more extensive immune response shortly after infection and none died from the virus.
“This study provides evidence in a living system, a real mouse, that the vaccines we devised can protect against the COVID virus using machine learning,” said MIT doctoral student Brandon Carter, one of the lead authors of the study. article, in an MIT news article. .
An interesting potential benefit is that immunocompromised people can get important protection from this approach, while the mRNA vaccines don’t work for them. Also “Long COVID” patients may get some relief in the form of a more extensive immune attack on their particularly resilient infection.
As the study summary puts it:
The undetectable specific antibody response in MIT-T-COVID immunized mice demonstrates that only specific T cell responses can effectively attenuate the pathogenesis of SARS-CoV-2 infection. Our results suggest that further investigation is worthwhile for panvariant T-cell vaccines, including for individuals unable to produce neutralizing antibodies or to help reduce lung COVID.
It’s a promising line of research and a great way to use advances in computation in the service of global health. But it’s also important to recognize that it’s still early days for the “pan variant” option. For starters, it can work with or against existing vaccines – what if the best peptides for the immune response vaccine are the ones targeted for destruction by mRNA priming? The two would work past each other. And too strong an immune response also carries the risk of collateral damage, accidental removal of ambiguously signaling cells and the like.
But these are good questions – questions that are relevant because the basic function of the new vaccine appears to be working. We’ll know more as the team runs more tests of this promising approach.