COVID-19 and the Path to Immunity

The emergence of adaptive immunity in response to the novel Betacoronavirus, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), occurs within the first 7 to 10 days of infection.1-3 Understanding the key features and evolution of B-cell– and T-cell–mediated adaptive immunity to SARS-CoV-2 is essential in forecasting coronavirus disease 2019 (COVID-19) outcomes and for developing effective strategies to control the pandemic. Ascertaining long-term B-cell and T-cell immunological memory against SARS-CoV-2 is also critical to understanding durable protection. A robust memory B-cell and plasmablast expansion is detected early in infection,2,4 with secretion of serum IgM and IgA antibodies by day 5 to 7 and IgG by day 7 to 10 from the onset of symptoms. In general, serum IgM and IgA titers decline after approximately 28 days (Figure), and IgG titers peak at approximately 49 days. Simultaneously, SARS-CoV-2 activates T cells in the first week of infection, and virus-specific memory CD4+ cells and CD8+ T cells reportedly peak within 2 weeks but remain detectable at lower levels for 100 or more days of observation. Grifoni et al1 and others5,6 have identified SARS-CoV-2–specific memory CD4+ T cells in up to 100% and CD8+ T cells in approximately 70% of patients recovering from COVID-19. Although severe COVID-19 is characterized by high-viral titers, dysregulated innate inflammatory cytokine and chemokine responses and prolonged lymphopenia, antibody-dependent enhancement or dominant CD4+ TH2-type cytokines (eg, IL-4, IL-5, IL-13) do not appear to contribute to acute COVID-19 severity. The magnitude of the antibody and T-cell responses can differ and be discordant among individuals and is influenced by disease severity (asymptomatic, mild, moderate, or severe). The immune correlates of protection are not yet defined for COVID-19, but neutralizing antibodies, especially those that recognize the viral receptor binding domain (RBD) and other epitopes on the spike protein that prevent subsequent angiotensin-converting enzyme II receptor binding, membrane fusion, and viral entry, is one path to immunity. The magnitude of the anti–SARS-CoV-2 IgG and IgA titers to the spike protein correlates in convalescing patients with CD4+ T-cell responses1; and the magnitude of IgG1 and IgG3 RBD enzyme-linked immunosorbent assay (ELISA) titers correlates strongly with viral neutralization.2,3 The generation of neutral izing antibodies directed at the spike protein is a basis of multiple human vaccines in clinical trials7 to counteract SARSCoV-2, and virus neutralization is the basis of potential therapeutic and preventive monoclonal antibodies also currently in human clinical trials. Such virus neutralizing antibodies are protective in animal models of SARS-CoV-2 infection. Potent neutralizing antibodies and TH1-biased CD4 + T-cell responses to the spike protein protect against SARS-CoV-2 infection in the lungs and nasal mucosa of nonhuman primates without evidence of immunopathological changes.8 The RBD region of the SARS-CoV-2 spike protein shows little sequence homology to the seasonal coronaviruses.2 Although variants in the SARS-CoV-2 spike protein (eg, D614G) may be a concern, SARS-CoV-2 has a low variation rate and so far, such variants have not been shown to reduce the recognition of RBD epitopes important for antibody neutralization. Recent reports have demonstrated a decline in IgG neutralizing antibodies to SARS-CoV-2 in convalescence, raising apprehension of susceptibility to reinfection. Antibody levels always decline after the acute phase of infection because most of the plasmablasts, the “effector” response of B cells, induced during the first weeks after infection are short-lived. A similar pattern is seen with the effector CD8+ T-cell response. After this reduction, serological memory is then maintained by the smaller number of long-lived plasma cells that reside in the bone marrow and constitutively secrete antibody in the absence of antigen. The antibody recall response comes from this pool of memory B cells that are also long-lived. In fact, rare circulating memory cells have been shown to produce highly potent neutralizing antibodies when serum neutralizing titers are low.3 Thus, an early decline of neutralizing antibody levels should not be of concern. The key is at what levels the antibody titers stabilize after natural infection or vaccination. This represents the generation of long-lived plasma cells to protect against subsequent infection. About one-third of patients who have recovered from COVID-19 do have antibodies with low anti-RBD titers and low viral neutralizing activity, especially among those who have had mild or asymptomatic disease. Given the wide range of clinical disease, this variability in the antibody responses among patients with COVID-19 is expected. Antigen burden is a major driver of the magnitude of the response, as notably the highest neutralizing antibody titers are found in severe disease,3 but other factors also could be involved. Based on models of immunity and reinfection dynamics of the common cold coronaviruses, human coronavirus (HCoV) OC43, HCoV 229E, and HCoV HKU1, limited protective immunity to SARS-CoV-2 has been suggested. In a study involving a human challenge to a circulating coronavirus (HCoV 229E), IgG and IgA antibodies waned over the first year after viral nasal challenge suggesting that protection against repeated infections with common cold coronaviruses lasts only 1 or 2 years.9 However, following experimental rechallenge with the same HCoV 229E strain at 1 year, no individuals who had been previously infected developed VIEWPOINT