SARS-CoV-2, the virus that causes COVID-19, is a new virus and we’re constently learning more about how the body responds to it. Scientists have rightly been urging caution in the interpretation of antibody test results because the tests have only been mainstream for a number of weeks. We’re still gathering data on how people produce antibodies in response to the virus and for how long these antibodies last. However, over the last few weeks some exciting data has been published in a number of studies which is beginning to shed some light on the situation.
What are antibodies again?
It’s probably worth having a quick refresher on what antibodies are and what they do. When your body is faced with an infection, such as a virus, it produces small defence molecules called antibodies which stick to the invader and help the immune system to kill it. There are different types of antibodies but the two main ones you’ve probably heard of to do with coronavirus are IgM and IgG. IgM is the bodies “rapid-response”; these antibodies are produced quickly - within a day or so of infection - but they’re often not well-targetted enough to clear the infection. As the immune system learns more about the invader, it produces more targetted IgG antibodies. This can take days to weeks in some people. These IgG antibodies are generally more effective than the IgM antibodies and help the body clear the infection.
After the initial infection, the body stops producing the IgM antibodies but it keeps on producing the IgG antibodies for some time. This gives the body immune system some “memory” so if it comes into contact with the same infection again it can usually be rapidly cleared without ever taking hold. This is what “immunity” is based on and it’s why most of us don’t get diseases like Chicken Pox twice. It’s also the how vaccination works.
Some people who get coronavirus don’t seem to produce antibodies we can detect.
Several studies have now been published showing that most people who have a confirmed infection with coronavirus produce antibodies. One chinese study of hospitalised patients with confirmed SARS-CoV-2 found that all 285 patients enrolled in the study became antibody positive within 20 days of their symptoms starting17. An unpublished study in France found that 159 out of 160 hospital workers with coronavirus developed IgG antibodies18.
However, other studies have found a small proportion of people who dont develop antibodies or who develop very low levels of antibodies in response to the infection. A study from China, which hasn’t yet been formally published, found that 90% of 1470 patients with confirmed SARS-CoV-2 developed IgG antibodies19. A non-peer reviewed study from the UK suggested that up to 8.5% of people did not develop IgG antibodies up to 60 days after SARS-CoV-2 infection20. There is some evidence that low or undetectable antibody responses are more likely in those with mild symptoms.
Antibody levels might decrease relatively fast.
Antibodies stay in the bloodstream for different amounts of time. For some infections, such as chickenpox, someone who was infected as a child will still have positive IgG antibodies many years later. However, for infections like SARS-CoV-2 we think antibodies are likely to last for a much shorter time. For example, we know that antibodies against SARS, a closely related virus to SARS-CoV-2, generally last for around 2-3 years21.
At the moment, we have limited information about how long antibodies to SARS-CoV-2 will last. However, an interesting study from China recently compared antibody responses in patients with and without symptoms who had confirmed SARS-CoV-2 infection. They found that about 80% of people developed antibodies, regardless of whether or not they experienced symptoms. However, people with symptoms tended to have higher levels of antibodies than those without symptoms. The scientitsts running the study then monitored the antibody levels over time and found that antibody levels dropped in 90% of people. In 40% of patients with no symptoms, antibody levels actually dropped to undetectable levels by 8 weeks. This only happened in about 12% of people who’d had symptoms.
What does this mean for me?
Based on these early data, we can begin to have some tentative thoughts about antibody responses to SARS-CoV-2. Probably around 90% of people who get infected with coronavirus will develop antibodies which are detectable by a blood test. The more ill you are with coronavirus, the more antibodies you’re likely to produce and the longer they’ll probably last. If you’ve not had any symptoms at all, then any antibodies you do produce may not last that long. This may mean that any immunity developed after the infection wont last for long either.
With regards to testing, it’s a good idea to get an antibody test fairly soon after you have recovered from any symptoms you think may be related to coronavirus - two to four weeks would be ideal. If you have a postive antibody test, you may want to get a repeat test in a month or so to see if you’re still positive.