Immune changes reflect severity of COVID infection

New research aims to predict who will get severe COVID from how their immune system responds from the start
18 January 2021

Interview with 

Ken Smith, University of Cambridge


An artist's interpretation of a coronavirus particle.


The coronavirus pandemic continues to paralyse countries across the world. The UK has nearly double the number of patients in hospital now compared with last March when the outbreak began, and last week saw over 1500 people die on a single day. The bizarre thing about COVID-19 is that some people get off scot free, while others have a very rough time and symptoms that persist for months after the virus has been cleared from the body. But could clues to why this happens - and to whom - be present in the way the immune system responds to the virus right at the start of the infection? Cambridge University immunologist Ken Smith thinks so. He and his colleagues have collected multiple samples from hundreds of COVID patients over the days to weeks as their individual disease syndromes progressed. And in those samples is a clear pattern that shows a very different immune response in people destined to get severe COVID compared with people who don’t. There are also hints in there of what “long covid” might be, as he told Chris Smith...

Ken - We recruited over 200 patients, ranging from people with very severe disease on intensive care, all the way through to healthcare workers who were screened positive for the virus while feeling perfectly well. And then we followed those patients up and kept taking samples at intervals out two to three months, to understand the evolution of the disease.

Chris - When you say samples, blood samples?

Ken - Yes. Patients who entered the studies all had to have a positive swab. So we had viral information at the beginning of the study, and then some patients we gathered that later on, but blood samples were the main thing that we studied.

Chris - What were you asking of those blood samples? What were you actually looking for and measuring?

Ken - The two key things we wanted to explore were how effective was the immune response, and the nature and the extent of inflammation that was coinciding with those immune changes.

Chris - Given that we've got this really broad spectrum of disease symptoms, the syndrome goes from people with no symptoms whatsoever to people who pass away from the infection. Do any of those symptom patterns map onto changes that you see in these blood samples in the patients you studied?

Ken - Yes. So we saw two key changes. The first thing was that in patients who had very mild disease or indeed no symptoms of disease at all, they had evidence of a more robust, very early immune response. In contrast, in people who ended up doing badly, even in the very first blood sample we took, which was often within a day or so of them first developing symptoms, there was already evidence of profound abnormalities in immune cell number, across a range of different cell types, and evidence of systemic inflammation. So they already had an abnormally inflammatory immune response, even at the very first blood test. So it could be [that] people are predisposed from the outset. And the implications of that is if we're going to act early to prevent this inflammation, we have to have very early. We can't wait and watch.

Chris - Are those changes present sufficiently early that if someone had their diagnosis in the community, they could be identified as at high risk or at low risk, right at the get-go? So that the intervention could be different. It could be tailored and therefore, potentially, the outcome for them could be changed.

Ken - That's what we have to aim for. What we don't know is when this inflammation starts. We know that it starts pretty well at the time symptoms develop. It may be that in fact, the abnormally inflammatory immune response is present even before symptoms develop. And how you detect such people is really of course difficult because they don't know they've got the infection yet.

Chris - And when people get better, do these changes all revert back to normal? Or are there persistent changes which could account for some of the longer term symptoms people complain of when they've had coronavirus?

Ken - I've only looked out to three months at this stage of our cohort. We are still following. So we will have six months data very soon. But at three months, the interesting thing is that there's still evidence of marked immune abnormalities in many patients. And those immune abnormalities can be seen even in patients who've got better and gone home. So those things could underpin some of the clinical features of long Covid. And so what's going to be important for us to do is to follow patients up at later time points to see if they in fact do resolve beyond three months, or if they represent an ongoing problem for patients.

Chris - And what shape do those abnormalities take in the wake of having had the infection previously?

Ken - So the immune cell abnormalities are largely a reduction in number of a lot of the different white cells in the blood, so-called lymphocytes, many different lymphocyte classes are profoundly reduced. And some of those cell numbers remain very low after three months, which could be consistent with, or could lead to an inability of good immune responsiveness to, for example, other infections or secondary infections. And that's why we have to see if those recover.


Add a comment