COVID-19: Acute and Convalescent Antibody Titers and their Implications for Herd Immunity and Vaccine Development (30 June 2020)

In this COVID-19-themed post, I briefly describe two recent studies that looked at SARS-CoV-2 antibody titers in patients who have been infected with the virus. By way of background, antibodies, also known as immunoglobulins, are Y-shaped proteins found in the serum and plasma portions of blood. They are mainly produced by plasma cells which, along with complement and antimicrobial peptides, comprise the “humoral” component of immunity. This is in contradistinction to “cellular” immunity, which is mediated by cells such as phagocytes and lymphocytes. Within humans and other mammals, there are five classes or “isotypes” of antibodies: Ig (for immunoglobulin) A, IgD, IgE, IgG, and IgM. (For more on the subject, see my 20 April post at: https://medium.com/@michaelzapor/covid-19-antibody-mediated-immunity-and-antibody-testing-20-april-2020-c42ca8d4d2f9) Here, I will only mention the IgG and IgM isotypes. IgG provides the majority of antibody-based immunity against invading pathogens, and its presence in circulation simply indicates prior exposure to a particular pathogen. In contrast, IgM is produced in the early stages of infection, before sufficient IgG is produced, and is a marker of recent infection. Although antibodies are made in response to most infections, they do not always confer protection against reinfection. For example, antibodies against Treponema pallidum, the bacterium that causes syphilis, do not necessarily prevent reinfection with T. pallidum. Additionally, even if antibodies are protective against a particular pathogen, levels may wane over time, eventually falling below protective titers. This is the basis for immunization schedules that call for revaccination in order to boost antibody titers against specific pathogens. For example, the Advisory Committee on Immunization Practices recommends a tetanus and diphtheria booster every ten years after the original series is complete (https://www.cdc.gov/vaccines/hcp/acip-recs/vacc-specific/dtap.html). Whether or not antibodies against SARS-CoV-2, the virus that causes COVID-19, are protective and whether or not protection is long-lived are two pressing questions that have implications for both the duration of the pandemic and for development of an effective vaccine. In this respect, scientists have hoped that SARS-CoV-2 would “behave” (anthropomorphically speaking) like two other coronaviruses, SARS-CoV-1 and MERS-CoV, both of which induce detectable antibody titers for at least eighteen months after infection (See: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4880093/) However, two recent studies suggest that this may not be the case.

The first of these studies, posted on the preprint server medRxiv on 16 June (https://www.medrxiv.org/content/10.1101/2020.06.13.20130252v1), looked at SARS-CoV-2 antibody titers between 29 February and 29 April 2020 in four groups of individuals in Wuhan, China: 1) 1,470 patients hospitalized with COVID-19, 2) 3,832 healthcare workers without known COVID-19 (but who were presumed by the study authors to have been infected at some point), 3) 19,555 “general workers”, and 4) 1,616 patients hospitalized for a diagnosis other than COVID-19. Among the nearly 1500 patients hospitalized with COVID-19, antibody testing conducted at least 21 days after symptom onset revealed detectable SARS-CoV-2 IgG titers in 89.8% (95% CI 88.2–91.3%); and among healthcare workers (who “inevitably” had been exposed at some point), IgG was detected in 4%, which was roughly comparable to levels in general workers (4.6%) and just slightly higher than levels in patients hospitalized with a non-COVID-19 diagnosis (1%). Similarly, IgM was detected in 31.4% of COVID-19 patients, 1.5% of healthcare providers, 1.3% of general workers, and 0.2% of non-COVID-19 patients. Because IgG and IgM levels were undetectable in ~10% and ~69%, respectively, of COVID-19 patients and because antibody titers were undetectable in most healthcare workers, the study authors concluded that infection with SARS-CoV-2 is “unlikely to produce long-lasting protective antibodies against this virus”.

In a letter published in Nature Medicine on 18 June, another group of Chinese researchers report the immune responses of 37 asymptomatic individuals and an equal number of symptomatic individuals who had been diagnosed by reverse transcriptase polymerase chain reaction (RT-PCR) amplification with COVID-19. (https://www.nature.com/articles/s41591-020-0965-6) Between the two groups, virus-specific IgG levels were significantly lower in asymptomatic individuals during the acute phase. Also, among asymptomatic individuals, 93.3% and 81.1% had a reduction in IgG and neutralizing antibody levels, respectively, during the early convalescent phase (i.e. within 2–3 months after infection), as compared to 96.8% and 62.2% of symptomatic patients. Moreover, 40% of asymptomatic and 12.9% of symptomatic patients became seronegative (i.e. had a loss of detectable antibody) in the early convalescent phase. According to the study authors, these data suggest that asymptomatic individuals mounted a weaker immune response to SARS-CoV-2, and that the early loss of protective antibody titers may have implications for “shield immunity” (i.e. the strategy of reintroducing immune persons back into society to slow the transmission of the virus), a term coined by Weitz et al. (See: https://www.nature.com/articles/s41591-020-0895-3)

Although the results of these two studies cast doubt about the longevity of immunity to SARS-CoV-2, they are not necessarily cause for despair. Both studies had potential confounding variables. For example, the first study included a cohort of healthcare workers who, according to its authors, were “inevitably” exposed at some point to the virus. However, according to University of Manchester immunologist Daniel Davis in an interview with Newsweek: “This is arguable because this assumes that (the) transmission rate of the virus was high in this hospital at that time, something which can’t easily be tested.” (https://www.the-scientist.com/news-opinion/studies-report-rapid-loss-of-covid-19-antibodies-67650) Although the results of both studies are provocative, the consensus among scientists seems to be that our understanding of SARS-CoV-2-associated immunity is incomplete, and that both time and larger study populations are needed in order to draw conclusions about the implications for herd immunity and vaccine development. Additionally, these studies looked exclusively at the humoral (i.e. antibody) response to infection with SARS-CoV-2, and as of yet, the role of cellular immunity is uncertain. Lastly, it’s worth noting that while natural infection (i.e. infection with the wild type pathogen) usually induces a more robust immune response than does a vaccine, this is not invariable. Examples of vaccines that induce a response that is more robust than wild type infection include the tetanus vaccine, the human papillomavirus (HPV) vaccine, the Haemophilus influenzae type b (Hib) vaccine, and the pneumococcal vaccine. (https://www.michigan.gov/documents/mdch/Waiver_Ed_Natural_Immunity_479884_7.pdf) Therefore, it may be possible that a well-engineered vaccine, including an appropriate adjuvant, might confer adequate protection against COVID-19.

As with my previous posts, my intention here is not to politicize, sensationalize, or trivialize the pandemic, but simply to provide information and, whenever possible, to alleviate anxiety.

Until my next update — regards.

Michael Zapor, MD, PhD, CTropMed, FACP, FIDSA

(30 June 2020)