By the time people recover from a viral infection, they have usually had an immune response and developed protection against the disease. That is, the immune system has produced antibodies that will recognize the virus if it attacks a second time, and those antibodies will know how to fight it off.

But a recent investigation at Schwabing Hospital in Munich, Germany, suggests this might not be the case for SARS-CoV-2, the coronavirus that causes the disease COVID-19. 

Clemens Wendtner, a chief physician at the hospital, tested COVID-19 patients for immunity after they had been treated for the disease at the end of January 2020. 

The tests showed a significant decrease in the number of antibodies. Wendtner says "neutralizing" antibodies, which stop a viral attack, fell in four out of nine of the patients who were tested, within two to three months.

Those findings coincide with a similar investigation done in China. That study also found that antibodies in COVID-19 patients do not persist in the blood. 

Further research is still required. But these initial findings suggest that a second infection is possible, where normally patients would have developed an immunity. And that may change the way experts handle things like the easing of social distancing measures. 

What is long-term immunity?

Our bodies develop immunity to the virus by producing antibodies. Antibodies are proteins that are produced by a type of immune cell, called B-cells. 

If a virus enters the body a second time, neutralizing antibodies catch the virus and stop it entering cells. And that stops the infection.

There are different types of antibodies. "IgM" is the first to form and its presence indicates an infection. As the infection clears, IgM antibodies decay and can no longer be detected. 

Once that has happened, "IgG" antibodies form and can persist in blood for months. The presence of IgG antibodies confirms a previous infection and a current immunity. 

B-cells mature and form plasma cells which reside in bone marrow, where they produce antibodies against infections we have had. Those plasma cells can persist for years — keeping us immune for years.

Read more: Coronavirus: German vaccine study draws thousands of volunteers

Antibody testing in COVID-19 patients

There are several ways to diagnose a SARS-CoV-2 infection. One way is a PCR test, which indicates the presence or absence of the virus by directly capturing its genetic material. 

Other tests aim to detect antibodies. Those tests provide indirect information about an infection. An anti-IgM positive test indicates a current infection, whereas an anti-IgG positive test indicates recovery and immunity. 

Mass antibody testing is valuable as it provides data on the immunity status of a community. Antibody tests can reveal COVID-19 cases which are asymptomatic or mild, and may have avoided diagnosis as a result.

If, however, SARS-CoV-2 antibodies do not last in blood, as those recent studies suggest, recovered patients have little protection and can be re-infected. 

Which would bring us back to square one, where everyone is at risk of infection. 

One way to stop a virus spreading in a population is by developing "herd immunity." That's when enough people have been infected, recovered and developed a lasting immunity. 

Some experts have said we may reach herd immunity with COVID-19 when about 60% of populations become immune to the novel coronavirus. But we can't be certain about that number just yet — with some viruses it can take up to 90% of a population to develop immunity before they reach herd immunity.

The benefit is clear: If we reach herd immunity, communities should be able to reopen, people should be able to return to their workplaces and resume their usual social activities. 

But now that secondary infections with SARS-CoV-2 seem possible, experts are questioning the easing of social distancing measures, or whether it's a good idea to issue recovered patients with so-called "immunity passports," allowing them to roam freely. 

Testing the effectiveness of immunity

With most viral infections, the size of an antibody response corresponds to the severity of the infection, and SARS-CoV-2 is no exception. 

Read more: The immune system's fight against the coronavirus

A Chinese study published in the scientific journal, Nature Medicine, suggests that asymptomatic patients have a weaker immune response to the infection. And asymptomatic patients lost antibodies more rapidly than patients who had had a more severe case of the illness. 

The study focused on 37 symptomatic and 37 asymptomatic SARS-CoV-2 patients. Its authors say that more than 90% of both groups showed a decrease in their neutralizing antibodies. 

Chinese researchers have also extracted antibodies from patients to test how effective they are in preventing the virus from entering human cells. The tests were done on cells in a laboratory, "in vitro." The researchers tested antibodies from 175 patients, and almost all protected cells from viral infection. 

But it is not yet known whether SARS-CoV-2 antibodies have the same effectiveness "in vivo" (inside the body) as they do in vitro. 

And scientists still don't know why those antibody levels decrease over time.

To compare: Antibodies against other types of coronavirus last in the blood for at least one year. That's true for SARS-CoV, the virus that caused a 2003 outbreak in Southeast Asia, and MERS-CoV, the virus that caused a 2012 outbreak in the Middle East. 

Antibodies and vaccines

All this has implications for the development of a vaccine against SARS-CoV-2. There are at least 130 candidate vaccines in clinical or pre-clinical trials worldwide. 

Conventional vaccines use weak or inactive versions of a virus to produce an immune response and protective antibodies. Newer types, known as DNA or RNA vaccines, hope to use genetic information from the virus to achieve the same result. 

But if natural antibodies decrease so quickly, how long will antibodies that are produced as a response to a vaccine last? As vaccines for SARS-CoV-2 have yet to be approved, we have a way to go before we find out.