How We Know Immunity Lasts – Whatever Imperial Says

One of the big stories yesterday was the latest data from Imperial’s REACT antibody survey showing that SARS-CoV-2 antibody levels in England have dropped by over a quarter in three months. The Telegraph has the details.

The mass research indicated that, by last month, fewer than one in 20 people had developed antibodies to Covid. Commissioned by the Department of Health, it is part of the largest piece of a research programme informing Government policies.

Its findings showed that by June, after the first wave of the pandemic, just 6% of the population had developed antibodies, which suggest some level of protection against the virus. Three months later, that figure had dropped to 4.4%, with most of the decline happening within just six weeks.

The sharpest fall was seen in those most in need of protection, with antibody levels among the over-75s reducing by close to 40% between June and September.

Scientists said the findings showed Britain is “miles off” achieving herd immunity, which they warned might never be reached without a vaccine.

However, the research did not examine the role played by other forms of immunity. Some scientists believe the part played by T-cells – a type of white blood cell that helps the immune system fight off viruses and is linked with prior infections by common colds – could be more crucial in fighting the virus.

Scientists analysed home fingerprick test samples from hundreds of thousands of adults to establish “detectable antibody levels” over a period of three months, and found levels fell by 26.5% overall.

The largest fall was among those most vulnerable to serious illness from Covid. Among those aged 75 and over, antibody levels fell by 39%, while a drop of only 15% was seen in those aged between 18 and 24.

The lead researcher is Helen Ward, a professor of public health at Imperial. She says:

I think what we are showing is that there is a really big challenge to that, which is that immunity is waning quite rapidly. After three months, we’ve already shown a 26% decline in antibodies. When you think that 95 out of 100 people are unlikely to be immune, and therefore likely to be susceptible, then we are a long, long way, from anything resembling a population level protection against transmission.

It is staggering that it is October and Prof Ward can still assert, largely unchallenged by journalists and politicians, that 95% of the country are susceptible. How is it that the lead researcher of a key Government antibody survey is still in the dark about long-lasting and pre-existing T-cell immunity, as Dr Mike Yeadon explains here, Peter Doshi in the BMJ explains here, and the CEBM explain here? There has been close to zero reinfection so far, and almost all the tiny number of documented cases of reinfection are mild, which as these immunologists explain, is exactly what we would expect, antibodies or not. Besides, when it comes to coronaviruses, “immunity” doesn’t necessarily mean never-get-it-again, it means the immune system is primed so even if a virus does breach some defences it is much less likely to breach all of them and cause serious illness.

Yet Professor Wendy Barclay, head of the Department of Infectious Disease at Imperial, agrees with her colleague, saying Britain is “miles off” herd immunity:

Seasonal coronaviruses that circulate every winter and cause common colds can reinfect people after six to 12 months – and we suspect that the way that the body reacts to infection with this new coronavirus is rather similar to that. We don’t yet know what level of antibody is needed in a person’s blood to protect them from infection or reinfection from SARS-CoV-2, but of course that level is a crucial thing to begin to understand. Most of the vaccine strategies are aiming to produce that level, and that level will feed into whether or not a population becomes immune or has any level of immunity.

It is odd that Prof Barclay claims that most of the vaccines are aiming to produce enough antibodies to protect from infection, as the trial protocols state otherwise. As Peter Doshi explains in the BMJ: “None of the trials currently under way are designed to detect a reduction in any serious outcome such as hospital admissions, use of intensive care, or deaths. Nor are the vaccines being studied to determine whether they can interrupt transmission of the virus.” All they are doing is testing to see whether mild symptoms become milder. That’s primarily because serious symptoms are so rare trials can’t detect enough of them to obtain reliable results. Yet the scientists at the heart of advising the Government appear to be unaware of this fact. What hope then for the politicians?

But back with declining antibodies, and Harry de Quetteville has written an extraordinarily misleading piece in the Telegraph, a classic of the genre. To keep it brief:

  • “With the four other coronaviruses that just cause us colds, antibodies generated by natural infection can be short-lived and we can suffer again as soon as six months later… Of course, the fact we repeatedly suffer from those four other coronaviruses suggests that natural T-cell response is unlikely to confer long-lasting immunity.” No, first because there is a huge variety of viruses which cause colds, not just coronaviruses, which is the main reason we keep getting them. And second, reinfection with the same virus is often mild because of immunity.
  • “Vaccines prompt our bodies to work in ways they wouldn’t do normally – that’s the point.” It really isn’t. Vaccines stimulate natural immune responses by simulating infection. Anything a vaccine can do will typically be done more effectively (if more riskily) by natural infection.
  • “The very prospect of declining natural immunity is devastating to those who want to ‘let the virus rip’. What if we did, thousands died, and yet a few months later there was no benefit, in terms of immunity?” Leaving aside the “let rip” slur – who’s Harry supposed to be quoting here, given that not a single advocate of focused protection has used that phrase? – declining antibodies are not necessarily declining immunity, and places like Sweden, London and New York are the elephants in the room for the supposedly unanswerable “what ifs”.
  • “This study is a reminder of what has been true about coping pre-vaccine from the beginning: to control transmission without draconian lockdowns, tests must be combined with effective contact tracing.” Yet no European country has run a successful test and trace programme. This is because, as the GBD FAQ explains, contact tracing “does not work for widely spread diseases such as annual influenza, pre-vaccine measles, COVID-19, or, by definition, against any pandemic”. Is Harry just reproducing a press release from Matt Hancock’s officer verbatim?
  • “We cannot be sure about durability, but even if it does not confer eternal immunity, a vaccine could still provide long enough immunity essentially to eliminate transmission, with breakout infections being rare, and, hopefully, as is often the case, less severe.” Then, suddenly, an acknowledgement that less than perfect immunity provides a decent level of personal and community resistance. But, oddly, only if it’s conferred by a vaccine rather than natural infection. Huh?

Stop Press: Prof Sunetra Gupta was on talkRADIO yesterday defending focused protection, herd immunity and the GBD in the light of the new panic around fading immunity. Watch it here. If anyone has time to rustle up a transcript for us we’ll publish it in a future update.