This is partly because we are measuring two different things – do the vaccines stop people getting infected and do they stop them getting severely ill? Prevention of infection is quicker to estimate because there are many more cases: these are the data were are seeing now. Reliable data on preventing severe illness and death are yet to emerge. But, as vaccines stop infection and stop severe illness in different ways, we cannot infer a vaccine’s effectiveness against illness or death from data on infection.
Infection begins when the virus spike protein binds to molecules on cells in our noses and throats. Imagine the spike protein is a key and the cell surface molecules are the lock. If the spike protein key fits the cellular lock, the door opens and the virus enters. The spike of the original COVID-19 virus wasn’t a great fit for our cells. It was good enough – with a bit of wriggling it could unlock enough cells to get a foothold - but it wasn’t ideal. The spikes of the variant viruses are a better fit: the lock glides open and virus floods in, so the variants are more infectious. Antibodies bind to the spike, blocking access to the lock; the door remains closed and the virus cannot infect. Unfortunately, antibodies induced by the original spike (the version in all current vaccines) don’t bind as tightly to the new spikes and some variant virus gets through. This is why we hear of positive COVID-19 tests in people who have been fully vaccinated. That’s the bad news.
The good news, however, is that prevention of severe illness does not rely on neutralising antibodies. Once the virus is inside the cell, antibodies are no longer effective. We now need the big bazookas of the immune system – killer T cells. Killer T cells drill holes in the wall of virus-infected cells and inject tiny packets of toxins. The infected cell shrivels up and dies or fills with water and explodes like a balloon, stopping the virus in its tracks. Crucially, T cells identify infected cells by detecting tiny pieces of virus protein on the cell’s surface. Almost any piece of any virus protein can be a flag for a T cell, and people who have been infected with or vaccinated against COVID-19 have hundreds of different varieties of T cells, recognising hundreds of different bits of the virus. Thankfully, the small number of mutations that allow the virus to evade antibodies have very little impact on the ability of this very diverse population of T cells to kill infected cells.
The greatest evolutionary pressure on the virus is its ability to replicate and move from host to host. If it can infect, replicate and move to a new host before the T cells show up, it will flourish. It must evolve to evade antibodies but there is much less evolutionary pressure to evade T cells. So the virus continues to circulate, but very few people will die. This is how the pandemic ends, eventually.
Eleanor Riley is professor of infectious disease immunology at the University of Edinburgh.