COMMENTARY

COVID-19: Testing Times for Everybody?

Prof Karol Sikora

Disclosures

April 21, 2020

This transcript has been edited for clarity.

Hello, it's Karol Sikora here, talking now about antibody testing. We at the Rutherford Cancer Centres, the network of four centres scattered around the UK, decided to test our staff for antibodies IgG [immunoglobulin G] and IgM [immunoglobulin M] against coronavirus.

Test, Test, Test

The logic really comes from Tedros Adhanom Ghebreyesus, the WHO Director General's concept of "test, test, test" being the most important thing we can do at the moment.

There are two types of test as you know. One is a PCR test for the presence of the virus. That's often called the antigen test, but it is done in a distant laboratory from the sample collection, which is usually a nose swab, or a throat swab. Lots of sample variation, lots of sample error there.

The second type of test is to look at a small sample of blood, either venous blood from venesection, or more likely a pinprick. Just 10 microlitres is all you need of blood or serum to do the assay and measure antibodies.

The virus [antigen test] and the antibody test tell you different things. The virus test tells you if the patient's been infected and is still infected.

The antibody test tells you that the patient at some point has been infected and has mounted an antibody response.

Test Kits

Now, we tested our staff. They're predominantly young and average age around 35. And we've tested now 120 of our staff and we're still doing it this week, using kits that come from Sugentech  in South Korea.

Now there's been a lot of noise about these kits not working. We validated our kits against the gold standard of antibody, a chemiluminescent assay, a much more cumbersome, laboratory-based test in North London.

All the staff at the centre here in Reading were tested. And there was complete concordance between the laboratory-based complex test and the kit test, which just takes a drop of blood and the answer's ready at the patient's side within 15 minutes.

Results

So, using this we went out and did the rest of the staff. The positive rate of either IgM, IgG or both is around 10%. Now at first I thought this is disappointing. Only 10% are seroconverted. What's the other 90%? Have they not had coronavirus?

And then I started looking around at the literature. The biggest study was in the town of Gangelt in Germany on the Dutch border. And there it was clear that having tested the whole population as much as possible, they found the conversion rate was only 14%. And yet, this was the epicentre of a pandemic of coronavirus a month ago. And yet, only 14% were producing measurable antibodies.

Sure, there's a lot of kits out there; there's a lot of variation. The price of the kits varies from $10, which are our ones, to $39. And no one really knows what the answer is.  

Problems

What we're looking for are two things. The first is information; information on how prevalent viral infection is and how good the immune response is to it.

The second thing we're looking for is some sort of passport, some sort of signal that a patient is safe, either to go back to work, or to travel on international airlines.

Unfortunately, my view, having looked at the data here, is that the antibody test will never do that. It will never confirm that someone has had the disease and is safe to go back. It's likely they've had the disease. But we don't know that it means that can't get the disease again.

The second problem we've got moving forward is that we need a simpler test. We need something that is done extremely simply, a saliva-based test would be ideal. A piece of litmus paper almost that you put in the mouth, and it comes out and after a minute it changes colour. And that's what a few imaginative biotech companies are now beginning to work on. That would be the future.

It probably won't be ready in 6 months. Public Health England who were responsible for testing in this country have been very slow, both to get off the mark with the virus testing by PCR, and, not only that, with antibody testing.

They seem to say no testing should be done without them. That's fine, well come and help us.

They couldn't even provide our validity lab with any control serum samples, we wanted 10 from positives, 10 from negatives: easy to get negatives, just go back in the fridge in the lab and identify 10 sera which have been stored since Christmas. At Christmas time there was no virus in the community, so they're definitely going to be negative. We needed 10 positives; we had to go out and get them ourselves from patients that our colleagues in London general practices and specialists had identified as having COVID-19 clinically. And that's been just tremendously helpful.  London's COVID-19 surveillance group of doctors has been responsible for this remarkable collection.

Looking Ahead

So how are we going to go forward with all this? We'll get more data, some very interesting data. I mentioned Gangelt in Germany. There's also cruise ship data coming in, and now a French aircraft carrier in which 2000 sailors, a pandemic proportion, was out off the coast of Brest; the ship came to the harbour and the sailors were tested, not with antibody tests but for the virus. Nearly 1200 of them were positive for the virus; encouragingly, only 600 actually had symptoms. We're waiting to see at what point they seroconvert, and does it go down the true roots of IgM and IgG?

So what's this all mean in terms of measuring? I think it's likely there are other mechanisms in play immunologically, other than the IgG and IgM. We know secreting IgA is one of the primitive ways of destroying outside pathogens in the mucosa before they actually get into the body. And it could be that the IgA is keeping the virus away from the conversion to IgM and IgG. All very complicated and it will take some years for the immunologists to sort this out.

I did a PhD a long time ago, 45 years ago on tumour immunology, and I remember this but I still have to go back to the books to sort it out. It's a very complex area. There are also cellular mechanisms that play NK [natural killer] cells, T-cells, cytokine-releasing cells, helper cells and so on, that identify the viruses as foreign and may mop it up, just gobble it up, long before there's any antibody response.

We're going to hear a lot about this. Public Health England have really not covered themselves in glory about the strategy they've used. My son is a manager in the Royal Mail and one of their big PCR testing services is coming from Belfast. It's a headache for the Royal Mail: it costs double to send a packet across the water, and not only that, the chances of delay and forgetting where the package has got to is much higher if you have to send it across the water.

I think what we're going to see now is a much better realisation that we've got to rationalise everything. We've got to be smarter about both the virus test and the antibody test.

Let's see how this all unfolds. Don't rush out and buy one for yourself and your family, it won't tell you anything you don't know. If you've had corona, great, especially if you've been asymptomatic. I hope I'm in that category. Let me know what you think. Thank you very much.

You can follow Karol Sikora on Twitter

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....