Infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) initiates a humoral immune response that produces antibodies against specific viral antigens such as the nucleocapsid (N) protein and spike (S) protein, which include specific anti-S protein antibodies that target the spike’s S1 protein subunit and receptor binding domains (RBD). Serologic tests can detect the presence of these antibodies in serum within days to weeks following acute infection. However, serologic testing should not be used to diagnose acute SARS-CoV-2 infection. Serologic tests can identify persons with resolving or past SARS-CoV-2 infection and thereby help scientists and public health experts better understand the epidemiology of SARS-CoV-2 individuals and populations at higher risk of infection. Although the immune correlates of protection are not fully understood, evidence indicates that antibody development following infection likely confers some degree of immunity from subsequent infection for at least 6 months. However, it is not known to what extent emerging viral variants may impact immunity from subsequent infection.
Nearly all immunocompetent persons develop an adaptive immune response following SARS-CoV-2 infection, including B and T cell-mediated immunity (1-3) due to antiviral humoral and cellular immune responses, respectively. Our understanding of the immune response to SARS-CoV-2 is incomplete but rapidly advancing. In humans, the humoral response includes antibodies directed against S and N proteins. The S protein contains two subunits, S1 and S2. The S1 subunit contains the RBD that mediates binding of virus to susceptible cells. RBD is the main target for neutralizing antibodies. Antibodies – including IgM, IgG, and IgA – against S and its subunits can be detected within 1-3 weeks after infection (4, 5). IgM and IgG antibodies can arise nearly simultaneously (4); however, IgM (and IgA) antibodies decay more rapidly than IgG (4, 6). The clinical significance of IgA in SARS-CoV-2 is not yet established.
How long anti-SARS-CoV-2 antibodies persist after infection remains unknown, although IgG antibodies, including IgG against the S and N proteins, persist for at least several months in most persons (7). Seroreversion has been reported among persons with mild disease (8). Persons with more severe disease appear to develop a more robust antibody response with IgM, IgG, and IgA all achieving higher titers and exhibiting longer persistence (8, 9). The observed persistence of antibodies can vary by assay (10), and some studies have found that approximately 5-10% do not develop detectable IgG antibodies following infection (11, 12). Although neutralizing antibodies may not be detected among patients with mild or asymptomatic disease (13), the humoral immune response appears to remain intact even with loss of specific antibodies over time (14). SARS-CoV-2 neutralizing antibodies that inhibit viral replication in vitro mainly target the RBD (2, 3). A need exists for standardized assays that can correlate antibody titers with neutralization (15).
SARS-CoV-2 reinfection has been documented (16, 17); however, studies indicate that persons with anti-SARS-CoV-2 antibodies are less likely to develop subsequent infection than persons without such antibodies. Outbreak investigations from a fishing vessel and a summer camp in the United States found that persons with pre-existing SARS-CoV-2 antibody were protected from subsequent infection (18, 19). In sequential outbreaks among staff and residents of two British nursing homes, persons who tested antibody-positive following the first outbreak were approximately 96% less likely to become infected during the second outbreak four months later (20). In a British prospective cohort study of persons with and without SARS-CoV-2 antibody, the adjusted incident rate ratio for subsequent infection was 0.11 among persons followed for a median of 200 days after a positive antibody test, compared to those who tested negative for anti-SARS-CoV-2 antibody (21). Another British cohort study found an 83% reduction in SARS-CoV-2 infection incidence over a five-month period among persons who had tested antibody positive for SARS-CoV-2 or had prior infection documented by revers transcription polymerase chain reaction (RT-PCR) (22). A large study in the United States of commercial laboratory results linked to medical claims data and electronic medical records found a 90% reduction in infection among persons with antibody compared to persons without (23), and another study of military recruits found that seropositive individuals had an 82% reduction in incidence of SARS-CoV-2 infection over a 6-week period (24). Additionally, antibody development following SARS-CoV-2 in humans infection correlates with a marked decrease in viral load in the respiratory tract, although a clinical correlation with viral load in the respiratory tract has not been definitively established (5). Experiments on non-human primates support the above observations in humans. Experimentally infected rhesus macaques that developed humoral and cellular immune responses were protected against reinfection when re-challenged 35 days later (25). Another study found that transfer of purified IgG from rhesus macaques infected with SARS-CoV-2 was effective in protecting naïve rhesus macaques from infection and the threshold titers for protection, based upon binding and neutralizing antibodies, were determined (26).
Taken together, the above findings in humans and non-human primates suggest SARS-CoV-2 infection and development of antibody can result in some level of protection against SARS-CoV-2 reinfection. The durability of this immunity has yet to be determined. While life-long immunity has not been observed with endemic seasonal coronaviruses (27), studies of persons infected with the novel SARS-CoV-1 and Middle East Respiratory Syndrome (MERS-CoV) coronaviruses demonstrated measurable antibody for 18 – 24 months following infection (28, 29), and neutralizing antibody was present for 34 months in a small study of MERS-infected patients (30). It is not known to what extent persons re-infected with SARS-CoV-2 might transmit infection to others or whether the clinical spectrum differs from that of primary infection.
SARS-CoV-2 infection begins when the RBD of the S protein of the virus binds to the angiotensin-converting enzyme 2 (ACE-2) receptor site in humans, the initial step in viral entry into human cells. Preventing SARS-CoV-2 from binding with ACE-2 receptors in the respiratory tract of humans can prevent infection and illness. This interaction between S protein of SARS-CoV-2 and the ACE-2 receptor sites has been the major focus of vaccine development. The vaccine candidates that have received EUA or are in late stage development aim to elicit neutralizing antibodies against the S protein or the RBD (31). Data from two phase III mRNA vaccine efficacy trials demonstrated up to 95% efficacy following a two-dose vaccination series (32, 33). It is unknown whether natural infection confers a similar degree of immunity compared to vaccination.
Natural SARS-CoV-2 infection results in antibody development against viral proteins including the N and S proteins, including the RBD of the S protein. Vaccine induced antibody development has implications for serologic testing. Before vaccine introduction, a SARS-CoV-2 serologic test that detects any of the N, S or RBD antibodies could be considered to indicate previous exposure to SARS-CoV-2. With the introduction of vaccine, vaccinated persons may test positive by serologic tests for the vaccine antigenic target (S and S subunits, including RBD) but not against other non-target proteins. Thus, history of vaccination and/or prior SARS-CoV-2 infection must be considered when interpreting serologic test results. Further, many persons infected with SARS-CoV-2 will be asymptomatic and never tested by viral detection tests, further complicating the interpretation of subsequent serologic testing. Testing for antibodies that indicate natural infection could be a useful public health tool as vaccination programs are implemented, provided the serologic tests are adequately validated to specifically detect antibodies to specific proteins (or antigens). Although an antibody test may employ a specific antigen(s), antibodies developed in response to different proteins may cross-react (i.e., the antigen(s) may detect antibodies it is not intended to detect), and therefore, may not provide sufficient information on the presence of antigen specific antibodies. For currently FDA authorized tests, it has not been established whether the antigen(s) employed by the test specifically detects only antibodies against that antigen and not others. Furthermore, none of the currently authorized tests have been specifically authorized to assess individuals who have received a vaccine. However, the EUA indications for currently authorized tests do not preclude the use of these tests on individuals who have received a SARS-CoV-2 vaccine. Vaccination may cause false-positive results for tests that utilize the S antigen or subunits like RBD, but not for tests that use the N antigen.
Considerations for public health and clinical practice
Accumulating evidence suggests that the presence of antibodies following natural infection may produce some level of protection from re-infection: 1) reduced incidence of infection among persons with SARS-CoV-2 antibodies followed for 3 months or longer; 2) data demonstrating that vaccination can reduce the incidence of illness (32, 33); 3) findings from outbreak investigations that pre-existing detectable antibody correlates with reduced incidence of infection (18, 19, 24, 34); 4) viral neutralization demonstrated with serum from persons following infection (2, 3); 5) decreased disease severity associated with administration of monoclonal antibodyexternal icon; and 6) challenge experiments with primates demonstrating prevention of re-infection (25). While it remains uncertain to what degree and for how long individuals with antibodies are protected against re-infection with SARS-CoV-2, or what concentration of antibodies may be needed to provide such protection, cohort studies indicate 80 – 90% reduction in incidence for at least 6 months among antibody positive persons (21-23). Longitudinal patient follow-up studies are ongoing to measure antibody levels before and after vaccination or natural infection to identify an association between responses below a certain threshold and vaccine failure or re-infection. These longitudinal patient follow-up studies are expected to elucidate the relationship between antibodies and protection from reinfection. In addition, T-cell-mediated adaptive immunity following natural infection, although not fully understood, likely contributes to protection from subsequent exposure to SARS-CoV-2 (35). It is also not known whether and to what extent viral evolution and the emergence of new viral variants may impact immunity from reinfection.