Coronavirus in Context: How Antibodies Work and What the Tests Measure

You're watching Coronavirus in Context. I'm Dr. John Whyte, Chief Medical Officer at WebMD. I'm joined today by Dr. Alan Wright. He's the Chief Medical Officer of Roche Diagnostics. Dr. Wright, thanks for joining me.

My pleasure.

We've been hearing so much about antibody testing. And there's a lot of confusion about what antibody testing does and doesn't do and who it's for. So I hope we have time to break it all down right now.

Let's start off with, for the audience, what is antibody testing? How does it work?

OK, well, let start with a little bit of the body's immune system and actually what happens when an invader comes into the body. So what happens is that an invader comes into the body, there is an immune response. And part of the immune response is the body creating antibodies.

Now, antibodies are proteins that attach to the invader. And the way to think about it, really, is that the antibody is like a sticky note that is stuck to the invader. And on that sticky note it says, kill me. And what happens then is these invaders are labeled with their sticky note, and then the white blood cells come and attack the invaders and destroy them.

Antibodies aren't initially available if you've never been exposed to the illness before. But what happens is after that first invasion, the body remembers. There are some antibodies and cells that make antibodies that linger in the body so that the next time the invader comes, the antibody can respond and attack again. That's what immunity is.

There are two ways to get immunity. One way is to have an infection. But the second way is to get a vaccine, which is part of the infection that doesn't cause an illness but still provokes those antibodies.

Tell us about the different types of antibodies, because that can be important as we think about protection.

So there are multiple different types of antibodies. And there are three main types of antibodies that are involved in the defense of the coronavirus, known as immunoglobulin A, immunoglobulin M, and immunoglobulin G. Immunoglobulin A lives in the mucous membranes. That's your eyes and in your mouth. And it is the first barrier to entry for an invader. Immunoglobulin M, think of that as the rapid responder. That's the first one that gets on the scene, typically not as effective as the specialist, which is a later-appearing immunoglobulin G, which appears much later in the infection and sometimes in the convalescent phase when you're actually getting better.

And there's different types of tests that we've been hearing about. There is point of care test. There are serum tests. Can you explain the differences between these two types of tests?

Sure, sure. Well, there are multiple types of point of care tests, all the way from the strip tests, which are dry chemistries which will detect the antibody in a strip-based system, very much like a pregnancy test that you're familiar with. Other point of care tests will run in a small device that's available in a clinic or a doctor's office. And then there are tests that run in very large labs. And these are very large analyzers capable of high capacity.

There is a tradeoff, though. There is a tradeoff either in precision and accuracy, how precise and accurate that test is, or speed. There are some point of care tests that are very convenient. The solid strip-based tests are very convenient, don't require a device. But there's a sacrifice in accuracy.

There are other tests that run on devices, small devices in clinics. But the sacrifice there is speed. And then the large analyzers, they run in laboratories. They are accurate and precise. But you need to get a sample to those devices.

There has been some data to suggest that some of the point of care tests-- because many of them are authorized, as opposed to approved by the FDA, requiring a lower standard of accuracy-- can be right half of the time. And that can give misinformation to consumers. What are you finding about some of the data relating to accuracy of point of care testing?

Again, it depends on the type of tests and the situation where that test is best used. Yes, there are less accurate, less precise point of care tests out there. But then there are other tests out there, point of care, particularly the molecular tests that are very precise and very accurate. But they're of low capacity, one or two samples at a time. And they take 15 to 20 minutes to generate a result. So it is very important to pair the test with what the purpose of the testing is and for the person who is interpreting the test to understand the precision and accuracy of that test.

Now, tell us what you're working on at Roche.

Well, Roche has several offerings that are either in the market, close to being launched in the market, or in development. So what we are working with that's already being used by the medical community are molecular tests. We have a very accurate, very precise molecular test that the text RNA-- which is the nuclear material, the blueprint of a virus-- and it is useful in detecting an active infection of the virus, or more precisely put, presence of virus in a person's oral pharynx.

The test that is soon to come to market is called a serology, or an antibody test. And that measures the immunoglobulins in a patient and is involved in looking at that immune response with the patient. And then in development, though development is a risky business and hard to put timelines, is certainly contemplating point of care testing.

The serology test requires a prescription, correct? Does the point of care as well?

All of our testing requires a prescription. And the prescription is because the results of the tests require interpretation by a physician and then development of a treatment plan for a physician.

Now, there's been debate, more so recently, that the presence of antibodies-- I'm just going to phrase it as most do. It has to be parsed. But the presence of antibodies may not necessarily mean protection. Some of this relates to qualitative reports, plus or minus, versus quantitative. Can you help provide some clarity on some of these blanket statements that say the presence of antibodies doesn't mean protection?

Well, first, I want to describe the evidence and our understanding of COVID-19 disease and the SARS CoV 2 virus. This is a new virus. We didn't know about this before December of last year. So we are watching the emergence of a new disease.

And that said, there is really no clinical correlates that we can work with with that disease. There was no body of evidence prior to December about this disease. So all the statements about immunity are based on extrapolations from other diseases. So in this case, common colds-- about 20% of the common colds are caused by coronaviruses. Not this coronavirus, but a virus in that family.

And some of those viruses really don't confer immunity over years like other viral diseases that we're familiar with, like measles or mumps or something like that. So that statement is really an extrapolation from our experience with other coronaviruses.

Now, the Rockefeller Foundation put out a report recently that we need to be doing 30 million diagnostic tests a week and that we also need to build capacity for antibody testing as well. Where do you see the role of antibody testing going over the next several months?

Antibody testing is very effective at quantifying the immunity of a population. And it's also effective at determining whether or not an individual has been exposed to a virus. Fortunately, the manufacturing capability of the large in vitro diagnostic manufacturers is substantial.

For example, when we launched our test in the beginning of middle of May, by the end of May our manufacturing capacity will be in the high double digit millions of tests. And all the manufacturers will have that same capacity. So the issue that we were hearing with the molecular testing, which has a much smaller infrastructure-- many fewer deployed analyzers in the market-- won't be the same with the serology testing.

And finally, what should we be telling patients about anybody testing when they're starting to think, should I get one?

Well, first, there are guidelines that define when antibody testing is necessary. I think the first thing is that as it stands now, antibody testing is not a diagnostic test for COVID-19 disease. Diagnosis is best performed with the correlation of clinical symptoms and a positive nucleic acid test, or a PCR test. So that is the first thing about antibody testing.

The second point with antibody testing is, as we know it now, antibody testing is very good at telling the patient and the treating physician whether or not that patient has been exposed to COVID-19 disease. Immunity, as I described before, will take more study and more understanding of the disease. But that's rapidly emerging.

Do you think reinvention can occur? The World Health Organization has said the presence of antibodies doesn't necessarily protect against reinfection. Or do you think those reinfections that we've heard about are failures of testing?

I don't know. You know, put it that way. I'm a true scientist. So one of the things about science is you're allowed to say I don't know.

But there are two causes of this reinfection story. One could be that in fact, they are reinfection. We don't know that. The evidence around the prior infection and then the reinfection is not well documented.

The second thing could be that it is a false positive or a false negative test. And that, again, gets to this discussion of accuracy and precision around testing and the fact that there is a great gradation of quality of testing solutions available in the marketplace. And it's very important for the treating physician to understand what tests that they're doing.

There's been a lot of studies recently-- many of them are in pre-print-- that had been using antibody testing to give us some sense of prevalence in a community, whether it's in California or elsewhere. Is that a good role for antibody testing?

Yes. Yeah, it's one of the strengths of antibody testing, is the determining exposure of a population to the disease. And what that will do is the information derived from that antibody testing will give better inputs to the epidemiologic models that we have and further refine our projections of the future course of this pandemic.

Well, Dr. Wright, I want to thank you, first of all, for that immunology lesson for all of us, as well as helping to explain where antibodies test in this fight against the global pandemic of COVID-19.

Thank you.

And thank you for watching Coronavirus in Context. I'm Dr. John Whyte.