What we actually know — and still don't know — about immunity to the coronavirus

• The coronavirus has infected more than 3.2 million people; more than 990,000 have recovered.
• Scientists are learning about what immunity to the virus looks like, but questions remain about how long it lasts and whether people can get reinfected.
• Visit Business Insider's homepage for more stories.

More than 990,000 people have recovered from the coronavirus, by official counts.

Typically, people who recover from infectious diseases, including other coronaviruses such as the common cold, develop immunity for some period of time. But there are still a lot of questions about what recovery means long-term for coronavirus patients.

"What we don't know yet is how long that immunity will last, the quality of that immunity, and whether all individuals will generate a long-lasting high immune response," Frances Lund, chair of the microbiology department at the University of Alabama at Birmingham, told Business Insider.

Here's what we know and don't know so far.

Are you immune to the coronavirus after you've had it?

Your immune system has a two-pronged way to attack the coronavirus, using the innate immune response and the adaptive immune system.

The innate immune system is your built-in, general immune system, which gets immediately activated against an unrecognized virus. This system includes physical barriers like skin, defense mechanisms like tears and bile, and cellular responses like inflammation.

The adaptive immune response is a powerful second layer: It has killer T-cells that destroy infected cells, macrophage cells that remove the leftover cellular debris, and B cells that produce antibodies.

Coronavirus antibodies — proteins that neutralize the virus based on its unique shape — should last in the bloodstream for a while, but experts aren't sure how long yet.

Case studies have found that most but not all recovered COVID-19 patients develop antibodies.

According to one report from Chinese scientists, which has not yet been peer-reviewed, about 10 of 175 participants studied (around 6%) did not develop antibodies. About 30% of those studied did not develop high levels of antibodies. The researchers also found that the levels of antibodies developed were correlated with age: Middle-aged and elderly patients tended to develop higher levels of antibodies than younger ones.

More research is needed on how the coronavirus interacts with the immune system, but these early findings suggest that some recovered patients might face a risk of reinfection.

How long does immunity last?

Dr. Anthony Fauci said in early April that people who recover from the coronavirus will likely be immune during a second wave of infection that's likely spread in the early fall.

"Generally we know with infections like this, that at least for a reasonable period of time, you're going to have antibodies that are going to be protective," Fauci said.

But what happens long-term is still unknown. A study from Chinese researchers on healthcare workers who recovered from SARS — which is also a coronavirus — in 2002 found that the number of SARS antibodies in their bloodstreams rose for the next few years, peaked in 2004, then declined afterwards until the study ended in 2015.

Although that research hasn't been peer-reviewed yet, it raises questions about whether recovered patients had "complete protection" from reinfection years later. Because the new coronavirus shares 79.5% of its genetic code with SARS, it's possible that antibodies for the new coronavirus might behave similarly.

There is also the possibility that if the virus mutates enough, immunity against one strain would not protect you from a new one, but Fauci said the coronavirus doesn't seem to be mutating enough to affect immunity in that way.

"If a person gets infected with coronavirus A, and then gets reinfected with a coronavirus, it may be coronavirus B," he said. "But right now, we don't think that this is mutating to the point of being very different."

Although the virus does change in subtle ways as it replicates and spreads, the small differences don't tend to affect how contagious the virus is or which symptoms manifest.

Have people already gotten reinfected?

Experts say that reinfection is unlikely.

The Korean Centers for Disease Control and Prevention reported earlier this month that more than 100 patients in South Korea retested positive for the virus after recovering and testing negative. Jeong Eun-kyeong, Director-General of the Korean CDC, said a new infection was unlikely, however, given that the follow-up tests were conducted within a "relatively short time" after the patients were released.

After closer investigation, the country's infectious-disease experts said on Thursday that the positive test results were likely caused by flaws in the testing process that led the tests to detect "dead virus fragments."

Dr. Oh Myoung-don, head of Seoul National University Hospital's division of infectious diseases, said the committee studying the cases found little reason to believe that the patients had been reinfected or that the virus had reactivated in their bodies.

There is also still uncertainty about a case reported in February in which a Japanese tour guide got the coronavirus, got better, then tested positive for it again three weeks later. Doctors aren't sure if she was reinfected or had not fully recovered from the first infection. (Plus, diagnostic tests can report false negatives and positives.)

How do I know if I have antibodies?

More than 100 healthcare companies are working to roll out tests that can detect antibodies in your blood.

Because infected people can be asymptomatic, mass antibody testing projects could offer public-health researchers a more complete picture of who's recovered from the virus. That would enable scientists to better estimate the percentage of the population that might be immune and calculate more accurate death rates.

Eight antibody tests have received emergency use authorization from the Food and Drug Administration so far. But companies are allowed to begin distributing their tests without that authorization as long as they provide disclaimers.

If you want to take a test if you suspect you've had the virus, you can ask your healthcare provider, find a lab in your area offering the test, or even order one online.

But these tests are still new and in the early stages of development, and studies have shown that they have varying levels of accuracy.

One report backed by the Chan Zuckerberg Biohub compared 12 tests and found that several companies are offering antibody tests for the novel coronavirus with less than 90% accuracy. That measure, known as sensitivity, ranged from 81.8% to 100%.

What does all of this mean for a future vaccine?

Scientists are racing to develop a vaccine for the coronavirus.

If exposure to the virus does not always produce an antibody response, there is concern that a vaccine might not always create immunity, either.

Still, an effective vaccine should work in enough individuals to build up herd immunity within the population.

To put the virus in decline, at least 50% of the population would have to be immune, perhaps far more. Only an estimated 2% to 3% of Americans have recovered from COVID-19 so far, though.