Why is it so hard to make a coronavirus antibody test? Scientists admit it can take YEARS to find the exact antibodies made by any infection - but say it's happening at 'lightning speed' for COVID-19

• Officials have pointed to blood antibody tests that indicate when someone has been infected with coronavirus as the best hope for 'reopening' the US 
• Blood tests show whether someone has developed proteins that will neutralize the virus if you come into contact with it again, giving you immunity
• Scientists admit it can take years to work out the exact antibodies produced 
• They also don't yet know how many antibodies are needed for immunity, or how long protection lasts. If it doesn't last a long time, a person could be re-infected 
• A Columbia University expert directing coronavirus testing at the school's lab breaks down what information scientists still need to create a reliable test  

Antibody tests for coronavirus that indicate precisely who has battled the infection have been pointed to as the key to reopening the US and Britain.

But despite various devices cropping up across the nation, government officials are yet to find one that is good enough for widespread use. 

A Columbia University expert admitted it could take years to find an antibody test accurate enough. 

But they added that scientists are desperately trying to find a blood-based test at 'lightning speed' - within a few months. 

Dr Susan Whittier said the challenge in making an antibody test lies with identifying the exact antibodies produced by the body to fight the infection.

It's important to understand how the virus itself triggers a healthy immune system to start producing antibodies, and scientists are divided on its behaviour.  

This results in antibody tests with various designs, some of which have failed to work with enough accuracy because they are not specific enough for SARS-CoV-2.

Doctors also say it is unclear, at this stage, the level of antibodies in the blood that provide immunity and how long that protection lasts. 

Some early animal studies suggested that antibodies could block reinfection for at least two weeks.  

But research on the new coronavirus' closest relative - SARS - shows a patient generates antibodies that last an average of two years.

Dr Whittier said the crux of the matter is, 'we don't know what we don't know'.

Labs across the US are producing antibody tests for coronavirus in the hopes of determining who has developed immunity - but their accuracy varies widely due to relaxed FDA regulation in the interest of expediting testing. Pictured: drive-thru antibody testing set up by USC 

HOW DO ANTIBODY TESTS WORK AND WHY IS IT DIFFICULT TO MAKE A RELIABLE ONE?  

The development of an antibody tests require some knowledge of the proteins that form the coat of the virus.

Viruses are made up of many proteins, called antigens, of which some are shared with other viruses. Only a few may be unique to the particular virus.

The proteins specific to the virus will trigger the production of antibodies that neutralise the virus, stopping it from replicating.

'We have to figure out what part of the virus is going to be really specific for that virus,' Dr Whittier, who heads up Columbia University and New York Presbyterian's microbiology lab, told DailyMail.com.

Those sections of the viral protein coat must then be produced in the laboratory, using cell lines, to be tested in an immunoassay.

Scientists 'take that protein, put it in the bottom of a plastic well and put the blood serum in it and see if there's something that will stick to it,' said Dr Whittier.

That 'something' would be the antibodies in the patients' blood.  

Anna Petherick, a lecturer in public policy at University of Oxford, said immunoassays will form the basis of home testing kits for people who think they have had COVID-19.

WHY IS IT DIFFICULT FOR SCIENTISTS TO MAKE AN ANTIBODY TEST SPECIFIC FOR SARS-COV-2? 

Coronaviruses invade cells through so-called 'spike' proteins, but those proteins take on different shapes in different coronaviruses. 

The University of Texas at Austin were able to map the spike proteins of SARS-CoV-2 in mid February - a breakthrough for developing a vaccine.

It was also a huge step forward for scientists developing an antibody test.

There are questions about which antigens (proteins) are best for an antibody test target, 'although the viral spike protein is universally perceived as the obvious candidate', Anna Petherick, a lecturer in public policy at University of Oxford, wrote in The Lancet.

Which part of the spike protein to use is less obvious, and is causing difficulty in creating a reliable antibody test.

'There is a lot hanging on the uniqueness of the spike protein,' Ms Petherick said.

'The more unique it is, the lower the odds of crossreactivity with other coronaviruses— false positives resulting from immunity to other coronaviruses. 

'The most similar of these is severe acute respiratory syndrome coronavirus (SARS-CoV), which led to the SARS outbreak of 2002.

'But another four coronaviruses cause the common cold, and ensuring there is no cross-reactivity to these is essential. '

A team at New York’s Icahn School of Medicine at Mount Sinai (NY, USA), has published details of antibody tests that use either the whole spike protein.

Others, such as Peng Zho at the Wuhan Institute of Virology in China, who was part of the team that sequenced SARS-CoV-2’s genetic code, have used the nucleocapsid protein and the spike protein.

'[The] nucleocapsid protein is the most abundant viral protein, which means it’s easy to detect. But we also chose spike protein because it’s very specific,' Dr Zho said. 'Actually, for coronavirus, the most divergent protein is [the] spike protein'But their development takes time,' she explained in The Lancet.

'Expressing the protein in the right structure is often the most difficult step.   

'In a nonnative system, such as a bacterial cell, the complex protein structures can come out slightly deformed, enough to stop antibodies from recognising them as they would the original viral coat protein. 

'There are also questions about which antigens (proteins) are best for this purpose. 

'Some diagnostic developers are cagey about giving away too many details, although the viral spike protein is universally perceived as the obvious candidate.'

Various labs making antibody tests might not even be testing for exactly the same antibodies. 

Some tests may confuse antibodies produced in response to the virus that causes COVID-19 to those made for other coronaviruses. 

'There are a lot of other coronaviruses, and the issue is you need to find what target is specific for this virus so it's not going to cross-react,' Dr Whittier said. 

Typically, finding the correct target would 'take months or years, and we're trying to do it in weeks to months,' Dr Whittier said. 

'We don't know the specific antigens or targets to look for.

'It seems to the lay public like it's taking a super long time, but from a lab perspective it's happening at lightning speed.'  

Labs are testing specific antigens by using the blood of patients who have been confirmed to have the infection. 

It will reveal if the antigen they have identified causes the antigens to stick. If it doesn't, it is not accurate. 

'At Columbia, we validated an antibody assay that was developed in Asia and tested lots of our [blood] serum that we had from patients and it turned out it was really specific - it only picked up SARS-CoV-2, which is good,' said Dr Whittier. 

'But it was only 50 per cent of patients who should have had antibodies.

'So if it was positive, that was good, they definitely have antibodies'.

But if it was negative, 'you might as well be flipping a coin'.

Needless to say, Columbia ditched that test. 

And with FDA guidelines relaxed in an effort to get more tests out more quickly, there's less assurance that validation is done with a comprehensive sample of patients. 

Dr Whittier says that the package insert for one test she looked at said the company had only tested their test on about five patients. 

'That's crazy,' she said. 'Normally that would never happen, but in the middle of a pandemic, you're allowed to push assays out because maybe perfect is the enemy of good.' 

+5

The blood of someone who's already had coronavirus will react with the strips on the test if they've developed antibodies - but different tests search for an immune response to different components of the virus 

ANTIBODIES PROVIDE IMMUNITY - BUT HOW MUCH PROTECTION AND HOW LONG IT LASTS VARIES  

When we contract an infection, the immune system goes to work creating specialized weapons against whatever invader we came into contact with, called antibodies. 

New Hampshire's test (pictured) looks for the presence of antibodies, but it may not tell what level of antibodies are in a person's blood, and scientists don't know how much is enough to offer protection from reinfection

Once we've encountered a pathogen and develop antibodies to it, these proteins sound the alarm when the invader returns and neutralize it. 

But not all antibodies are created equal, and not everyone develops the same number of antibodies. 

For example, it's well known that once you get chicken pox, you're almost certainly immune to it and will never be infected again. 

That's not true for antibodies against other pathogens. Immunity for other infections wears off relatively quickly. 

Flu is fairly well understood, but the virus has many strains which mutate readily. 

Antibodies produced against each variation of flu we encounter are quite specific to that unique infection. 

So when we come into contact with an evolved or different strain of flu the next season, the antibodies we developed the prior year don't do us much good. 

That's why flu vaccines are 'recombinant' - they're made based on a combinations of several strains of flu, triggering the production of a variety of antibodies to block the strains scientists think we might making their way around the globe that year. 

The most common coronaviruses - those that cause seasonal colds - trigger fairly weak antibody responses, lasting only a couple of weeks, which is part of the reason you might get multiple colds in a single year. 

HOW DO 'STRIP' BLOOD TESTS FOR CORONAVIRUS WORK?

Simple blood tests for coronavirus, like Premier Biotech's, work much like pregnancy tests. 

After the sample of blood is collected, a technician injects it into the analysis device - which is about the size of an Apple TV or Roku remote - along with some buffer, and waits about 10 minutes. 

The blood droplet and buffer soak into the absorbent strip of paper enclosed in the plastic collection device. 

Blood naturally seeps along the strip, which is dyed at three points: one for each of two types of antibodies, and a third control line. 

The strip is marked 'IgM' and 'IgG', for immunoglobulins M and G. Each of these are types of antibodies that the body produces in response to a late- or early-stage infection. 

Along each strip, the antibodies themselves are printed in combination with gold, which react when the either the antigen - or pathogen, in this case, the virus that causes COVID-19 - or the antibody to fight are present.

Results are displayed in a similar fashion to those of an at-home pregnancy test. 

One line - the top, control strip - means negative. 

Two lines - the top control line and the bottom IgM line - in a spread-out configuration means the sample contains antibodies that the body starts making shortly after infection. 

Two lines closer - control and IgG - together mean the person is positive for the later-stage antibodies. 

Three lines mean the patient is positive for both types of antibodies

However, research on the new coronavirus's closest relative - SARS - is somewhat more encouraging. By the second week after someone is infected, they've generated antibodies that seems to last an average of two years. 

But we simply don't know how similarly antibodies for the virus that causes COVID-19 will behave because we've only known it existed for four months. 

WHAT DOES A POSITIVE CORONAVIRUS ANTIBODY TEST REALLY MEAN? EVEN THE EXPERTS DON'T KNOW... YET 

Time and volume of people infected are two key crucial ingredients for an antibody test. They tell scientists how many antibodies are enough to make someone immune to reinfection, and how long that immunity lasts. 

And labs developing antibody tests have neither on their side. 

'We can't tell you that, because we don't have a gold standard to compare it to,' Dr Whittier said.

The FDA gave emergency use authorization to the first antibody test for coronavirus in the US on April 2 - less than two weeks ago. 

That's about as long as scientists think that it takes for a patient to mount an antibody response to SARS-CoV-2. 

Dr Whittier said: 'Twelve to 14 days is when most individuals are having an antibody response, but we don't know if it's protective, and we don't know how long it lasts.'

Having the antibody test is the first step to answering those questions. But some people will develop antibodies more quickly than other, and some will develop greater quantities of antibodies than others. 

It will take following these people and testing them repeatedly to learn what the 'gold standard' for immunity is. 

What's more, the first antibody tests only returned results about whether antibodies were present. They did not reveal what volume someone's body had produced them. 

Now, labs are starting to produce 'semi-quantitative' tests, that can tell if someone has 'a little antibody or a lot of antibody,' Dr Whittier said. 

As more people are tested for levels of antibodies, not just their existence, epidemiologists can study what levels provide protection and for how long.

But for now, 'we don't know what we don't know,' Dr Whittier says.