Long COVID: Developing a diagnostic test

Why is it so difficult to definitively diagnose long COVID?
16 August 2021

Interview with 

Danny Altmann, Imperial College London


a blood sample on a slide in a lab


One of the difficulties in studying and treating long COVID is that we don’t currently have a definitive way to diagnose it. And without a diagnostic test, the condition becomes much harder to manage or study. Eva Higginbotham spoke to Danny Altmann, an immunologist from Imperial College London where he’s been looking to see if there are any similarities in the blood of people suffering from long COVID...

Danny - Well, this follows on from the kind of ideas that the Akiko was talking about just now, that if you start from the assumption that long COVID means that something has gone awry in the long term in the immune system as a consequence of having been infected with SARS-CoV-2, there are many things one could look for, many signatures one could look for in blood. One of them is this thing of autoimmunity, the idea that the body has started attacking itself. So if you can work out a signature of antibodies against self, autoantibodies, that long COVID people tend to have in their blood that others don't you're on the way to having a test.

Eva - And so does that mean you're taking blood samples and then looking in those for specific signs of these antibodies?

Danny - That's exactly what we're doing. We're working with cohorts as widely as we possibly can. People who were hospitalised, people who were not hospitalised, people who didn't even manage to access a PCR test but who now have that clear set of persistent long COVID symptoms. And we're looking at the array of antibodies in their blood and working out which parts of the human proteome, the human protein antigens, they respond to so that we can work out what that test could look like and make something that could be accessible that could get them that diagnosis that they need.

Eva - And what have you managed to find so far?

Danny - Well, so it's fairly early days, we're trying to expand our early data out further as we go, as are many other labs around the world, including Akiko's at Yale. And so far we've been working especially with people where we've got quite a good handle on their long COVID from things like MRI analysis. And we have been finding some signatures that are reproducibly found in people with persistent symptoms that aren't found in those rapid recovery people. So we feel we're well on the way to having the kind of test that we want.

Eva - Do you think this kind of test, if you are looking for autoantibodies and those are causing the disease, could account for all the different kinds of long COVID that we see? Because there's such an array of symptoms.

Danny - Yeah, well, you know, long COVID is a very diverse and sort of scary business, isn't it, with so many different possibilities. And in medicine, yes, sometimes we like to stratify things as much as we can and show all the differences. For the purposes of this initial analysis, we've kind of gone to the other extreme and said, let's be completely agnostic about all the different things that long COVID could be and say, if we simply put together our long COVID groups and our rapid recovery groups and our healthy control groups, are there any blood markers we can find that are useful? And so far we feel like we're winning, but you know, I accept that down the line, you may want to say, well, could this be useful for telling the difference between a skin rash long COVID person and a brain fog long COVID person? And I'm sure it could be

Eva - And you mentioned all these different groups, is one of the groups you're looking at the people who are actually asymptomatic for COVID, but now seem to have developed long COVID afterwards?

Danny - Very much so. So to me, they're one of the most interesting groups, but they're also one of the most... everybody's desperate, but they're particularly desperate, aren't they? Because if you have long COVID - bear in mind this isn't obviously just a UK problem, it's a world problem - so we're talking about something like 20 to 40 million people on the planet with long COVID, all trying to access care, all kinds of convince their doctors they have a problem, and all trying to get into that care pathway. Imagine how much harder that is if not only do you not have a diagnostic test for your long COVID, you can't even prove to them you had COVID in the first place because you were asymptomatic, and perhaps people in your household had it or whatever. So, you know, I'm especially interested in those people.

Eva - And finally, if you did manage to make this kind of test, how do you think it could be used? Do you think it could be used to help predict who's most at risk of having long COVID for a really long time?

Danny - I think there are many nuances to be brought out of it, but in the first instance, I'd like it to be within that battery of tests that your doctor can order as for other autoimmune diseases like diabetes or arthritis, where your local hospital can just do the test and say, yes, tick the box, this is long COVID, of they go to the long COVID clinic, you know, let's try and help this person and do something for them.


Add a comment