Correlates of protection from SARS-CoV-2 infection.

Since the beginning of the COVID-19 pandemic, many scientists and public health officials assumed that infection with SARS-CoV-2 would protect from reinfection and that neutralising antibodies would correlate with protection or would be at least one of the protective immune mechanisms. Early on, these assumptions were supported by non-human primate data showing protection from reinfection, a correlation between neutralising antibodies protection, and protection afforded by passive transfer of neutralising antibodies.2, 3, 4 However, reports of rare reinfections, the notion that antibody titres might wane within weeks (which is incorrect), and the fact that human coronaviruses that cause common colds do cause reinfections have cast doubt on these initial assumptions.5, 6, 7

A study in the UK reported in The Lancet by Victoria Hall and colleagues, called the SARS-CoV-2 Immunity and Reinfection Evaluation (SIREN) study, suggests that being seropositive to SARS-CoV-2 through natural infection protects robustly from asymptomatic and symptomatic reinfection. The study analysed data from 25 661 enrolled health-care workers between June 18, 2020, and Jan 11, 2021, including 8278 individuals with known previous SARS-CoV-2 infection, of whom the vast majority were antibody positive at enrolment and 17 383 individuals who were seronegative and had not previously been infected with SARS-CoV-2. 21 617 (84·2%) of 25 661 participants were women and 4010 (15·6%) were men, with a median age of 45·7 years. 87·3% of the participants were White, 6·9% were mixed race, 2·0% were Asian, 1·6% were Black, 1·3% were Chinese, 0·6% were from other ethnic groups, and 0·2% preferred not to specify. Individuals were followed up with questionnaires (every 2 weeks), PCR for SARS-CoV-2 (every 2 weeks), and serology (at enrolment and every 4 weeks). 1704 infections occurred in the naive cohort, while two probable (required supportive serological data or supportive viral genomic data) and 153 possible (two positive PCR results 90 days apart or an antibody-positive individual with new positive PCR test 4 weeks after the first antibody positive test) infections occurred in the SARS-CoV-2-experienced cohort. Additionally, 864 individuals in the naive group seroconverted over the study interval but were not counted towards SARS-CoV-2 infections. The authors report previous SARS-CoV-2 infection provided a 84% risk reduction for reinfection (adjusted incidence rate ratio [aIRR] 0·159, 95% CI 0·13–0·19) and 93% risk reduction for those with symptomatic infections (aIRR 0·074, 0·06–0·10). Importantly, a variant of concern known as B.1.1.7 did circulate during the final part of the observation period, causing about 50% of all infections, but did not seem to have an effect on reinfection rates.

The findings of the authors suggest that infection and the development of an antibody response provides protection similar to or even better than currently used SARS-CoV-2 vaccines. Although antibodies induced by SARS-CoV-2 infection are more variable and often lower in titre than antibody responses induced after vaccination, this observation does make sense considering current SARS-CoV-2 vaccines induce systemic immune responses to spike proteins while natural infection also induces mucosal immune responses and immune responses against the many other open reading frames encoded by the approximately 29 900 nucleotides of SARS-CoV-2. The SIREN study adds to a growing number of studies, which demonstrate that infection does protect against reinfection, and probably in an antibody-dependent manner.9, 10, 11, 12, 13, 14, 15

The SIREN study does have several limitations. Different serological assay platforms were used to determine seropositivity and not all of them have the same sensitivity over time or focus on the spike of SARS-CoV-2, which is the prime target of neutralising antibody responses. Additionally, although protection against B.1.1.7 was shown, the degree to which infection with so-called garden variety SARS-CoV-2 provides protection against reinfection with antigenically distinct variants of concern such as B.1.351 and P.1 remains unclear. Furthermore, the authors did not link quantitative antibody measurements to protection. This is an important piece of the puzzle since the protective titre for SARS-CoV-2 is still unknown, although non-human primate studies suggest that it is likely to be low.2, 3 Establishment of antibody titres as a correlate of protection and defining a protective titre would be extremely important for public health considerations and for patient management. A correlate of protection and a protective threshold would also allow for the development of additional SARS-CoV-2 vaccines based on small immunogenicity-based phase 3 trials rather than large and costly field efficacy trials, which are becoming exceedingly difficult to perform. Determination of a protective titre should be a priority for future studies that investigate protection afforded by natural infection or vaccination.