New COVID-19 antibody test should tell us how we become immune
As the first wave of the COVID-19 epidemic sweeps across the world with more than 400,000 lives lost, the need to know whether a person has been infected or remains at risk is rising. A worldwide lack of credible antibody tests catalysed Denmark’s largest hospital, Rigshospitalet and the University of Copenhagen in collaboration with Denmark’s largest company Novo Nordisk A/S to join forces to develop an antibody test. This led to the development of not just one but two very reliable tests, one of which may also answer the question of how humans become immune to COVID-19.
Like a spear, the virus shoots its spike protein arm into the cell wall, where it attaches itself to the antenna that is intended to receive signals to regulate the body’s blood pressure. The virus is now attached and ready to be ingested by the cell and trick it into making thousands of copies of itself to invade other cells in the body. The way SARS-CoV-2 attacks the human body is well known, but what happens next is still very uncertain. Some people become seriously ill; others do not. Some develop antibodies; others may not. Some become immune; others may not. Antibody testing must now answer the many questions that need to be answered to avert successive waves of the COVID-19 pandemic.
“The overall goal of the test is to ensure that we are self-sufficient here in Denmark and that we know every detail about the test. That way, we are sure that it has high quality and can change the test if the virus starts to mutate and that we can monitor responses to a vaccine when that becomes available. But we have also developed the test further, so we can show in detail what type of antibodies are formed against the virus, when they disappear and measure which responses actually result in immunity,” explains Peter Garred, Professor of Clinical Molecular Medicine at Rigshospitalet and the University of Copenhagen.
Rapid tests not good enough
The new project started when it was realized that there were difficulties in sourcing a good COVID-19 test for Denmark. Together with pharmaceutical company Novo Nordisk A/S, Peter Garred and his team began developing a Danish antibody test from scratch to ensure that the entire population of Denmark can be tested.
“We know that when global epidemics – pandemics – hit the world, then there will be limited capacity, and there will be a fight for all these commercially available tests. So it is critically important for a nation like Denmark to be able to have unlimited capacity to test its population,” explains Mads Krogsgaard Thomsen, Executive Vice President, Head of R&D and Chief Science Officer, Novo Nordisk A/S.
The major challenge when developing an antibody test is choosing the right reagents. Many existing rapid tests do not appear to be good enough. They are built up like pregnancy tests and are not sensitive enough and may have some problems with specificity.
They might be used for surveillance of populations to some degree, but generally the tests produce too many false-negatives, and false-positives are even worse since they incorrectly tell people that they have been infected.
“I will not say they are useless, but they are of limited value. They do not always pick up the positive and the negative. So I would shy away from them,” says Mads Krogsgaard Thomsen.
A really good candidate
The first step in developing a sensitive and specific antibody test is to determine which characteristics of the virus you want to test for antibodies. When SARS-CoV-2 infects a person’s cell, the virus attaches itself to the surface of the cell using a spike protein that is inserted into the cell membrane’s ACE2 receptor.
“Once the virus is attached, the cell absorbs it and is then tricked into mass-producing many more viruses that can invade new cells. The infection has started. The spike protein is the protein that is crucial for the virus to become infectious. If it was not for the spike protein, the virus would no longer be infectious. So because it is so important for the virus, you can be quite certain that the spike protein will remain intact even when the virus mutates,” explains Peter Garred.
The immune system stores information about external threats: for example, the spike protein’s unique shape, so in a later attack the body can quickly mobilize its defensive antibodies, which tag virus-infected cells so that they can be destroyed.
“You have seven different coronaviruses. You want to make sure that what you measure is antibodies against this virus, the one called SARS-CoV-2, and not all the others. So you have to have something that is specific – something that does not really mutate since it is critical for the infection of the virus. Something that is specific for this particular virus and not found in a similar shape and form among other coronaviruses,” explains Mads Krogsgaard Thomsen.
In antibody terms, this is called an antigen, and the spike protein appeared to be a really good candidate to use for the test. However, this is also a tricky protein, since it is very big.
“That is actually a little bit difficult to express, because it is a very special protein that has some sugars on top of it. And if you don’t get them expressed, if you don’t produce them together with the protein in the right way, shape and form, then the antigen will not react in the right way with the antibodies; it will not pick up the antibodies. Part of the spike protein, the receptor-binding domain (RBD), which makes the virus become infectious, is very characteristic for SARS-CoV-2 compared with other coronaviruses and is easier to produce than the whole protein. Thus we choose to express that part in a human cell line to mimic what is going on in the host as much as possible,” says Mads Krogsgaard Thomsen.
5,000 people a day
However, it takes more than a good antigen to make a reliable test. The researchers used a sandwich ELISA test. Plastic wells are coated with viral antigen – the RBD domain of the spike protein – and the serum from the patient is added. If this serum contains antibody, it will bind.
“Only antibodies developed specifically against the RBD domain of the spike protein will bind. The plates are then washed and we add in the same antigen, but then this antigen is coupled to an enzyme that can bind to the other arm of the antibody. This can be developed to produce a chemical reaction and make a colour. Then we can see whether there is an antibody reaction against the virus,” explains Peter Garred.
Once the test has been developed, the final challenge is to scale up the testing. Laboratory technicians may be highly skilled in conducting a test assay, but automation is required to test a population of almost 6 million people or the whole world.
“So we have used robotic experts and wonderful nerdy people who can transform all of this into high-throughput format. Today we actually have the equipment to run the tests without any human intervention. At the end of the day, a computer picks up all the results and can send them back to the professor, who then can then identify those who were positive and send patients the right information,” says Mads Krogsgaard Thomsen.
The final trials of the test have just ended, and the results are so convincing that it is possible to start testing up to 5,000 people a day. However, a decision on how the new test will be utilized has yet to be taken.
“We have been able to test out all kinds of blood sera from minimally to maximally infected patients and identify that it picks up with 98% sensitivity the fact that you actually are positive, and with 99% specificity, meaning the risk of actually having a false-positive that you think you have had infection, but you did not have the infection,” explains Mads Krogsgaard Thomsen.
Do antibodies make you immune?
The overall goal of the project has been to become self-sufficient in Denmark. The researchers have therefore chosen reagents for the tests that are always easily available.
“We know that there has been a lack of reagents. We know that there has been a lack of all the different tools that can be used, and medicine for instance. I think what we have learned from this COVID-19 pandemic is that you have to rely a little bit on your own resources. And this means that you need to have a supply chain that works. It has been very important that we can make our own tests. Not only antibody tests but also other types of tests. So we can help the population,” says Peter Garred.
Due to the severity of the COVID-19 pandemic, having definite knowledge about the antigens that the tests use and about the quality of the tests is crucial.
“If the virus mutates in the RBD domain, we could also very easily change the antibody test by following how the virus develops, but also that we can use the test to monitor responses to the vaccine when that becomes available. And finally, that we can also monitor whether individuals build up good antibody responses,” adds Peter Garred.
The new test has to be able to basically detect antibodies both among people who are severely ill in respirators and those without symptoms, but one important question remains unanswered: do the antibodies make people immune?
“Well, this is something that we are not 100% sure about at this moment. We hope that the antibodies made by the immune system can neutralize the virus, so we become immune and cannot be infected again, but this is something that needs more research, and we do not know whether the antibodies indeed neutralize the virus, or whether you need a certain amount of antibodies or how long the immune response will last,” says Peter Garred.
A new test will investigate vaccines
The researchers suspect that immunity against COVID-19 may decline over time, and then people will become susceptible again.
“We know from common cold viruses, which are similar to the COVID-19 virus, that the antibody responses may decline and then we will have a new common cold infection later with the same virus,” explains Peter Garred.
The researchers assume that people with high levels of antibodies against the key elements of the virus will be more immune than others. To answer this question, the researchers have developed a second antibody test – a direct ELISA that can be used to investigate the type of antibody against the COVID-19 virus in the blood. Immunoglobulins M, A and G are found in various body fluids and confer different kinds of immunity.
“While the sandwich ELISA test is ideal for tracking the spread of infection in the community, this direct test can used to investigate the type of antibody over time. Once our knowledge of COVID-19 has improved, our tests will hopefully also be used to predict who has actually become immune and to select donors whose antibodies can be used to treat other COVID-19 patients,” says Peter Garred.
Further development of the second test will even enable the interaction between the spike protein of the virus and the ACE2 receptor to be studied to determine whether antibodies can inhibit the binding to the receptor, which enables the virus to become infectious.
We can thus investigate whether vaccines against the COVID-19 virus actually provide a protective antibody response and measure how this will change over time. Putting all these aspects together will provide a picture on how the immune system of each individual reacts towards the virus.
Avoiding the impact of a second wave
The collaboration between Rigshospitalet, the University of Copenhagen and Novo Nordisk A/S has been a perfect match. Novo Nordisk A/S produced the high-quality reagents for the antibody tests, contributed to developing a thorough test and ensured that the tests can be made available on a large scale. The researchers at Rigshospitalet and the University of Copenhagen have mainly done the molecular groundwork, tested prototypes and examined clinical samples.
“Public–private partnerships are critical. The public cannot produce a cure; the public cannot produce a vaccine; the public cannot produce this high-throughput testing that Novo Nordisk has done. But we, as companies, cannot take care of patients. We do not have the hospitals and the patient insights that they do. So there has been a high amount of mutual trust. So it has been a seamless collaboration, and this has been done in a matter of two months,” explains Mads Krogsgaard Thomsen.
There are no plans to commercialize the tests. Instead Novo Nordisk is currently evaluating how the tests can help on a larger scale, both within Denmark and internationally since, according to Mads Krogsgaard Thomsen, testing populations over and over again to fight the COVID-19 pandemic will be critically important.
“You want to follow the development of the epidemic both at the population level and also at the individual level, because you would like to know whether you have had an infection. So can you go and visit your old grandmother, because you have had it and you are immune and you will not infect her. And if you have had lung disease, and do not have antibodies, then it is important to be cautious. The tests can ensure safer reopening of societies to avoid the hardest impact of a potential second wave,” concludes Mads Krogsgaard Thomsen.
The new antibody test has been developed from scratch in close collaboration between Rigshospitalet, the University of Copenhagen and Novo Nordisk A/S, with support from the Carlsberg Foundation. The Novo Nordisk Foundation awarded Peter Garred a grant of DKK 4,803,750 for the project The Copenhagen SARS-CoV-2 Antibody Testing Initiative.