Skip to main content

Two Studies Show COVID-19 Antibodies Persist for Months

Posted on by Dr. Francis Collins

Antibodies against SARS-CoV-2
Caption: Artistic rendering of SARS-CoV-2 virus (orange) covered with antibodies (white), generated by an immune B cell (gray) at the bottom left. Credit: iStock/selvanegra

More than 8 million people in the United States have now tested positive for COVID-19. For those who’ve recovered, many wonder if fending off SARS-CoV-2—the coronavirus that causes COVID-19—one time means their immune systems will protect them from reinfection. And, if so, how long will this “acquired immunity” last?

The early data brought hope that acquired immunity was possible. But some subsequent studies have suggested that immune protection might be short-lived. Though more research is needed, the results of two recent studies, published in the journal Science Immunology, support the early data and provide greater insight into the nature of the human immune response to this coronavirus [1,2].

The new findings show that people who survive a COVID-19 infection continue to produce protective antibodies against key parts of the virus for at least three to four months after developing their first symptoms. In contrast, some other antibody types decline more quickly. The findings offer hope that people infected with the virus will have some lasting antibody protection against re-infection, though for how long still remains to be determined.

In one of the two studies, partly funded by NIH, researchers led by Richelle Charles, Massachusetts General Hospital, Boston, sought a more detailed understanding of antibody responses following infection with SARS-CoV-2. To get a closer look, they enrolled 343 patients, most of whom had severe COVID-19 requiring hospitalization. They examined their antibody responses for up to 122 days after symptoms developed and compared them to antibodies in more than 1,500 blood samples collected before the pandemic began.

The researchers characterized the development of three types of antibodies in the blood samples. The first type was immunoglobulin G (IgG), which has the potential to confer sustained immunity. The second type was immunoglobulin A (IgA), which protects against infection on the body’s mucosal surfaces, such as those found in the respiratory and gastrointestinal tracts, and are found in high levels in tears, mucus, and other bodily secretions. The third type is immunoglobulin M (IgM), which the body produces first when fighting an infection.

They found that all three types were present by about 12 days after infection. IgA and IgM antibodies were short-lived against the spike protein that crowns SARS-CoV-2, vanishing within about two months.

The good news is that the longer-lasting IgG antibodies persisted in these same patients for up to four months, which is as long as the researchers were able to look. Levels of those IgG antibodies also served as an indicator for the presence of protective antibodies capable of neutralizing SARS-CoV-2 in the lab. Even better, that ability didn’t decline in the 75 days after the onset of symptoms. While longer-term study is needed, the findings lend support to evidence that protective antibody responses against the novel virus do persist.

The other study came to very similar conclusions. The team, led by Jennifer Gommerman and Anne-Claude Gingras, University of Toronto, Canada, profiled the same three types of antibody responses against the SARS-CoV-2 spike protein, They created the profiles using both blood and saliva taken from 439 people, not all of whom required hospitalization, who had developed COVID-19 symptoms from 3 to 115 days prior. The team then compared antibody profiles of the COVID-19 patients to those of people negative for COVID-19.

The researchers found that the antibodies against SARS-CoV-2 were readily detected in blood and saliva. IgG levels peaked about two weeks to one month after infection, and then remained stable for more than three months. Similar to the Boston team, the Canadian group saw IgA and IgM antibody levels drop rapidly.

The findings suggest that antibody tests can serve as an important tool for tracking the spread of SARS-CoV-2 through our communities. Unlike tests for the virus itself, antibody tests provide a means to detect infections that occurred sometime in the past, including those that may have been asymptomatic. The findings from the Canadian team further suggest that tests of IgG antibodies in saliva may be a convenient way to track a person’s acquired immunity to COVID-19.

Because IgA and IgM antibodies decline more quickly, testing for these different antibody types also could help to distinguish between an infection within the last two months and one that more likely occurred even earlier. Such details are important for filling in gaps in our understanding COVID-19 infections and tracking their spread in our communities.

Still, there are rare reports of individuals who survived one bout with COVID-19 and were infected with a different SARS-CoV-2 strain a few weeks later [3]. The infrequency of such reports, however, suggests that acquired immunity after SARS-CoV-2 infection is generally protective.

There remain many open questions, and answering them will require conducting larger studies with greater diversity of COVID-19 survivors. So, I’m pleased to note that the NIH’s National Cancer Institute (NCI) recently launched the NCI Serological Sciences Network for COVID19 (SeroNet), now the nation’s largest coordinated effort to characterize the immune response to COVID-19 [4].

The network was established using funds from an emergency Congressional appropriation of more than $300 million to develop, validate, improve, and implement antibody testing for COVID-19 and related technologies. With help from this network and ongoing research around the world, a clearer picture will emerge of acquired immunity that will help to control future outbreaks of COVID-19.


[1] Persistence and decay of human antibody responses to the receptor binding domain of SARS-CoV-2 spike protein in COVID-19 patients. Iyer AS, Jones FK, Nodoushani A, Ryan ET, Harris JB, Charles RC, et al. Sci Immunol. 2020 Oct 8;5(52):eabe0367.

[2] Persistence of serum and saliva antibody responses to SARS-CoV-2 spike antigens in COVID-19 patients. Isho B, Abe KT, Zuo M, Durocher Y, McGeer AJ, Gommerman JL, Gingras AC, et al. Sci Immunol. 2020 Oct 8;5(52):eabe5511.

[3] What reinfections mean for COVID-19. Iwasaki A. Lancet Infect Dis, 2020 October 12. [Epub ahead of print]

[4] NIH to launch the Serological Sciences Network for COVID-19, announce grant and contract awardees. National Institutes of Health. 2020 October 8.


Coronavirus (COVID-19) (NIH)

Charles Lab (Massachusetts General Hospital, Boston)

Gingras Lab (University of Toronto, Canada)

Jennifer Gommerman (University of Toronto, Canada)

NCI Serological Sciences Network for COVID-19 (SeroNet) (National Cancer Institute/NIH)

NIH Support: National Institute of Allergy and Infectious Diseases; National Institute of General Medical Sciences; National Cancer Institute


  • Joseph says:

    They left out the T cell response.

  • Lisa Phillip Rimland says:

    Thank you for posting, and for your clear explanations of these studies. As our understanding of Covid increases, so does our hope for the future.

  • R.obert F.ranklin Smith says:

    Thank you! This is very timely (and very welcome) news. I do hope with this funding our NIH will continue with longer term antibody research (18-30 months seems appropriate). Best wishes, Rob Smith

    • M Dye says:

      Do you know if the NIH has continued their research into lasting immunity after Covid-19 infection? I’m finding it difficult to find any info post 11 months.

  • Martha Price says:

    Back-tracing evidence as well as visits to my physician’s office for symptoms, indicate that I developed COVID-19 symptoms on or about December 23, 2019, having been exposed to the virus on December 17. I was not supported to be tested for antibodies until May 20, at which time I tested positive via the Mayo Clinic lab. I re-tested on July 10, (then seven months after developing symptoms), via my local hospital lab, and again tested positive. At the end of September, I tested at LabCorp, and the result was negative. There have been many strange neurological symptoms, all but three which have finally subsided. Cardiac arrhythmia developed in December on the very same day that I developed a dry cough, and palpitations still continue, though less frequently. Documenting my symptoms daily since December, I eagerly read the research reports regarding all aspects of Covid-19.

  • Byron Mui says:

    This is good news. Since reinfection with SARS-COV-2 is rarely reported, it would be interesting to see if the T-cell immune response does play a role to prevent reinfection.

  • Michael N. Alexander says:

    The article suggests that monitoring IgG antibodies may be useful in tracking Covid-19. Are the antibodies in question unique to Covid-19, or could other conditions provoke their appearance?

  • Billy A. says:

    I tested positive june 7 until june 30.. It’s now Oct. but still have loss of taste and started getting weak very fast. Dr – Left ventricle enlarged, white, red, thyroids, platelets all very good.. Trying to find out why loosing energy and blood pressure dropping out of sight ?? 57 and not overweight, very fit but gaining weight. Drs have nothing so hoping to find something on this website to start me down some path to what’s going on…

  • John C. says:

    I tested positive for Cov March 15 2020. Lasted about 10 days with a low grade fever along with being super tired. 5 days of antibiotics helped the symptoms and my recovery. Took an antibodies test Oct 20 2020 and it came back positive for the antibodies from a Cov infection.
    It’s hard to describe lasting symptoms because these are not normal psychological times however I do have some brain fog some days and get tired. Working out helps and resting when tired is the best medicine.

  • Jay T says:

    My question and comment in regards to how many antibodies, especially IgG (long term or lasting antibodies), are produced per each antigenic covid-19 virus. What is the ratio of the production or replication of natural antibodies to each virus? 100 to 1? 1,000 to 1? or is it 10,000, or even 100,000, or more; ensuring adequate circulating antibodies, actual immunity, for any return bio-identical virus in the future to the same host. Decades ago physiologic studies stated that a healthy human male produced on average 400 million sperm on ejaculation for the purpose of a single sperm penetrating a single human female egg (ovum).Was perpetuation of our specie ever that desperate?Would the enormity of sperm production to perpetuate our specie, also be repeated in antibody production in an innate motive of protecting the life of a single member of our specie, the individual host, a covid-19 patient, if in fact Covid19 virus is lethal? In the cases of lifetime immunity with childhood diseases, e.g., of measles, mumps, and chicken pox, wouldn’t there be also an innate motive for lifetime immunity in the case of life threatening Covid-19 infection? Since upper 90 percentile of people contracting Covid-19 virus result in zero to mild symptoms, could the Covid-19 virus be just another seasonal flu and not lethal, instead the blame for lethality be placed on the patients’ pre-existing conditions, plus the added burden of the toxicity of their prescription drugs? Should this be anywhere near the case, or even otherwise, could the the Covid-19 virus itself actually be non-lethal?

  • TW Stewart says:

    Can this finding help with the development of a vaccine?

  • abishek says:

    The information is useful …

  • Tony Ross says:

    I can’t help but wonder if regular exposure to the virus would extend immunity for those who experienced symptomatic covid and recovered? If immunity results from exposure, ie from the body’s response, would it decrease if repeatedly exposed following the initial exposure? Why would regularly exposing those who have recovered with controlled amounts of the virus be a risk?

    • Mrs. C says:

      I love what you have said! Once there are memory cells they’ll be activated upon future encounters serving as a “booster”. Now, too much exposure and the inmune system will come to believe it’s “self” and ignore it (that’s the principle behind immunotolerance). I just think we should all get rid of the masks, especially if you’ve recovered from the disease or after a couple of weeks from vaccination.

      • Jim B says:

        I sometimes wonder how much the national belief in COVID DEATH is furthered by a media that has shown itself to be a willing accomplice in spreading ‘news’ that seems more consistent with a specific political agenda. That is why I find it so refreshing to come to a site like NIH and find actual discourse. Thank you all for sharing. So much is gained by being able to read multiple personal ‘takes’ on what has happened before, during and after infection and ultimate cure. Of course those who were not cured cant share and we must rely on their caretakers for information. Hope we see some more of THEM. Thanks, Be well.

      • Tony R. says:

        So, my wife and I just got another antibody test and it was positive again. To recap we had covid back in early July and tested positive for antibodies in October. Then in late December my sister in law came to stay with us because her kids tested positive. She took a rapid test and it was negative so between that and the fact that we had had it we felt comfortable in having her stay. She was with us for five days during which we we were in close contact constantly. She coughed occasionally but she has bad lungs and has coughs regularly so we weren’t too concerned. Then my wife and I went away for the weekend. On our way back Sunday we got a call from her sister telling us her cough had got worse and she developed a fever so she got retested and it was positive. My wife and I never developed symptoms but got tested ASAP just to be sure we were clear. We were. My point is we spent days in close quarters with someone who was highly infectious and nothing. And now we tested positive for antibodies. Re-exposure no doubt didnt hurt.

  • William H. says:

    Going on 5 months now of positive Antibodies. Getting tested every month at my expense as a personal study. We will see how long they last!

  • Colleen Sohmer says:

    As a 23 year veteran RN in NYC and a Covid19 survivor, this is so encouraging to hear! I’ve done my own monitoring with my antibodies and I’m still positive for IgG antibodies after 8 months. I’ve tested for antibodies 3 separate times since May of 2020. First symptom 3/16/20 with 1st PCR test negative on 3/30/20. Symptoms seemed to improve after a week or so then became worse again with PCR testing positive 4/8/20. I continued to have symptoms through mid May & June.
    My first positive IgG antibody test was on May 1st, then again in August and now November 4th it was positive again. So I’d say that’s promising!

    • nick D says:

      Had covid early March. Antibody positive in May blood test. Had blood work in November with another antibody test. Still positive for the antibody.

      • Tony R. says:

        Wife and I had it in the beginning of July. Tested positive for antibodies 2 weeks ago. Evidence is mounting that antibodies last and prevent reinfect ion in vast majority of cases. Also, number of confirmed of infections is minute and infinitesimal as a percentage of known cases, forget unknown cases. Obviously, no guarantees, but I like the odds. I do wonder why they don’t develop some sort of regular re-exposure protocol for those who have had it in order to maintain levels of antibodies, T cells etc. Like a nasal spray containing the virus or form of it. A nasal booster vaccine of sorts …

      • F. says:

        I also had covid in March. I just done another antibody test. Still positive. I’m trying to see what it means for me.

  • Carolyn says:

    Had Covid symptoms in Early February and took a Antibody test in Oct 20/20 and tested positive for the Antibodies 8 months later

    • michael n. says:

      I had a mild case in early March with no complications short or long term. I tested + for ABs at 3, 6 and now 9.5 months. I’m 84 years old and will be eligible soon for vaccine. I plan to take the first shot but seriously question the necessity or even safety of a “booster” 3 or 4 weeks later. Don’t my persisting antibodies (including T cells) make the first vaccine shot, in effect, the booster and if I take a second shot, might that cause an exaggerated immune response, even a cytokine storm?

  • Ian says:

    I lost my smell and taste 1 April 2020 and tested my blood on the 23 Dec 2020 to donate plasma this is the response 9 months later -from my Blood Bank – When we tested the level of COVID-19 antibodies in your blood we are able to confirm that the antibody level is at a mid-range which could potentially be used as a form of treatment to treat people who are ill with the disease. This result implies that you have been infected with or exposed to COVID-19.

  • Chicago Kid says:

    I had most symptoms of COVID in early February 2020, and I was quite ill. This was before the lockdowns, and when they were telling everyone that we were all safe unless you traveled overseas. I was never tested at that time–because I hadn’t traveled, the COVID test wasn’t offered, and so it was assumed I had the seasonal flu.

    Fast forward to May 1. I took the Quest Labs anitbody test as soon as it was available (3 months after my iacute illness). It as a serum blood test–negative. Then, in early November 2020, I took another test through the Northwestern University SCAN study (which looks for much lower levels of the IgG antibody than the commercially available serum COVID tests do), and it showed that I had the antibody “present.”

    My questions for the researchers: Why can’t you standardize the tests? Shouldn’t ANY amount of antibody present count as being “positive”? There shouldn’t be a threshold to meet, as in the commercially available tests. I’m sure there are a lot of people who had the illness who were told that they didn’t have the antibody because the tests were looking for levels of antibody that would only be found in acutely ill, hospitalized patients–but not in people who weren’t hospitalized and recovered on their own. My antibody test showed “positive” 9 months after I was sick. So which antibody tests can we actually trust for accuracy?

  • Edward G. says:

    I was detected in April 15th of 20 and cleared Mar 15 and had presence of antibodies at that time. To monitor my primary took another sample 11/23 and there was still the presence of the antibodies. I received a blood test 12/28 and just found out the antibodies are still present. Seems exceptionally rare.

  • Russe says:

    My wife had Covid in March and still had antibodies when tested at the end is December.

  • 1 2 3

Leave a Reply to M DyeCancel reply