Issue #51: COVID-19 vaccines and influenza vaccines: Part 5 – correlates of immune protection

Written by Nobel Laureate Professor Peter Doherty

Why is it that vaccines to prevent the development of symptomatic COVID-19 infection seem, when compared with the long-term influenza experience, to be working so well? We’ve explored that issue in depth from a number of aspects over the past few weeks (#48, #49, #50). But, though we have a long history with flu, much of the in-depth analysis of what COVID-19 vaccines are doing within us seems currently to be either ‘in progress’ or still in the planning phase. We assume that vaccine-induced antibodies – the immunoglobulins (Igs #21) – are the primary players in protection, but there is much to be learned re the details of how and where they are doing that job.

The question that the regulatory agencies, like the FDA in the USA and our TGA (Therapeutic Goods Administration) were asking of the vaccine manufacturers was: do they protect against the development of clinical symptoms? That has been answered in the affirmative, especially for severe disease. Evidence of protection has come through very clearly from both rigorously monitored clinical trials and from field reports of efficacy as these vaccines have been given to millions of people in, particularly, the UK, the USA and Israel. The latter are less closely analysed, but the numbers of people involved are so much higher that this data provides great reassurance as, in Australia, we launch our mass vaccination program somewhat later.

From that overseas experience, we have also learned that there are a few categories of people who may be at risk of severe side effects. This came through first for the PfizerBioNTech (BP) mRNA vaccine with the demonstration that people who are highly allergic (EpiPen users) could be at risk of anaphylaxis (#43). With the AstraZeneca (Az) vaccine, the concerns raised in Europe re blood clotting seem to be narrowing down to, especially, women under 55 with a history of DIC (disseminated intravascular coagulation) or CVST (cerebral venous sinus thrombosis), both rare conditions. It’s still early days but, by now, any clinic administering the vaccine will presumably be well aware of these suggestive findings.

What do we know about the immune responses resulting from the intramuscular (IM) injection of the BP and AZ vaccines? The information I’ve been able to access so far is a bit fragmentary but, especially after the second booster shot, both seem to be inducing protective, neutralising antibody responses. Neutralising antibody? With COVID-19, that’s IgM, IgG or IgA molecules (#21, #22) that block the RBD (receptor binding domain) on the SARS-CoV-2 spike protein and stop it from attaching to the ACE2 molecule that is expressed on many of our cells. Failure of attachment means that the virus cannot get into the cell, un-coat to release its RNA information package and begin the process of reproducing (#42, #43). The comparable interaction for those flu viruses that are prominent in causing infections that spread from person to person is that the viral hemagglutinin (HA) attaches to cell-surface sugars, sialic acids (especially the SAa2,6 form in humans) and is then interiorised. The SAa2,6 is widely expressed on many cell types in the upper and lower respiratory tract, though not on the terminal alveoli of the lung where gaseous exchange (air to red blood cells) occurs.

When it comes to preventing any infection using a non-living (like the BP and AZ products) vaccine (#47) that is administered into the arm, the key protective molecule will be virus-specific IgG, the main form that we find in blood for a fully developed immune response. While early IgM and, particularly, IgA production (#21, #22) is now known to be prominent in limiting the consequences of SARS-CoV-2 infection in an immunologically-naïve person (previously uninfected or vaccinated) these molecules are, in people receiving the BP or AZ vaccines by IM administration, less likely than IgG to be present in the mucus bathing the nasal epithelium. Though I haven’t yet seen data on IgG levels in the noses of vaccinated people, it’s particularly gratifying that both the BP and the AZ vaccines seem to be reducing transmission.

In addition, if there is any virus production at site of entry in vaccinated people, once that gets going and causes some local damage it’s likely that more IgG will spill over from blood to limit further virus spread, both within us and to others. And, though the vaccines in use so far are not designed to prime the CD8+ ‘killer’ T cell response that eliminates the virus-producing cell ‘factories’ (#33, #34), they do have some limited capacity to do that. More detailed analysis via the many clinical trials looking at immune responses in vaccinated people across the planet – we have several under way in Melbourne – will no doubt tell us that.

Especially in Australia where we have had a very limited exposure to infection, those trials could tell us an enormous amount about both this disease, and about vaccine-induced immunity in general. Here, the only known cross-reactivity is with some elements of the response to the common cold coronaviruses. Still, there is much less of a background than we would find for any ‘novel’, pandemic influenza virus, where everyone who has had influenza will have established, cross-reactive immune memory in both the T cell and B cell (#18) compartments.