Humoral immunity to SARS-CoV-2 and seasonal coronaviruses in children and adults in north-eastern France

Abstract

Background: Children are underrepresented in the COVID-19 pandemic and often experience milder disease than adolescents and adults. Reduced severity is possibly due to recent and more frequent seasonal human coronaviruses (HCoV) infections. We assessed the seroprevalence of SARS-CoV-2 and seasonal HCoV specific antibodies in a large cohort in north-eastern France.

Methods: In this cross-sectional seroprevalence study, serum samples were collected from children and adults requiring hospital admission for non-COVID-19 between February and August 2020. Antibody responses to SARS-CoV-2 and seasonal HCoV (229E, HKU1, NL63, OC43) were assessed using a bead-based multiplex assay, Luciferase-Linked ImmunoSorbent Assay, and a pseudotype neutralisation assay.

Findings: In 2,408 individuals, seroprevalence of SARS-CoV-2-specific antibodies was 7-8% with three different immunoassays. Antibody levels to seasonal HCoV increased substantially up to the age of 10. Antibody responses in SARS-CoV-2 seropositive individuals were lowest in adults 18-30 years. In SARS-CoV-2 seronegative individuals, we observed cross-reactivity between antibodies to the four HCoV and SARS-CoV-2 Spike. In contrast to other antibodies to SARS-CoV-2, specific antibodies to sub-unit 2 of Spike (S2) in seronegative samples were highest in children. Upon infection with SARS-CoV-2, antibody levels to Spike of betacoronavirus OC43 increased across the whole age spectrum. No SARS-CoV-2 seropositive individuals with low levels of antibodies to seasonal HCoV were observed.

Interpretation: Our findings underline significant cross-reactivity between antibodies to SARS-CoV-2 and seasonal HCoV, but provide no significant evidence for cross-protective immunity to SARS-CoV-2 infection due to a recent seasonal HCoV infection. In particular, across all age groups we did not observe SARS-CoV-2 infected individuals with low levels of antibodies to seasonal HCoV.

Funding: This work was supported by the « URGENCE COVID-19 » fundraising campaign of Institut Pasteur, by the French Government's Investissement d'Avenir program, Laboratoire d'Excellence Integrative Biology of Emerging Infectious Diseases (Grant No. ANR-10-LABX-62-IBEID), and by the REACTing (Research & Action Emerging Infectious Diseases), and by the RECOVER project funded by the European Union's Horizon 2020 research and innovation programme under grant agreement No. 101003589, and by a grant from LabEx IBEID (ANR-10-LABX-62-IBEID).

Keywords: COVID-19; SARS-CoV-2; antibody response; seasonal coronaviruses; sero-epidemiology; seroprevalence.

Fig. 1

Epidemiological context and description of sampling (n = 2544). (a, b) Collection of samples occurred during the first wave of COVID-19 in France. (c) Majority of sera were drawn from children below 20. (b,d) Samples were collected from 11 hospitals located predominantly in north-eastern France.

Fig. 2

Antibody concentrations and neutralising activity of (a-d) pre-pandemic samples (n = 90) and (e-h) samples collected in 2020 (n = 2449). The dark red dashed line in panels c and g is the cut-off defining seropositivity for the Luciferase-Linked ImmunoSorbent Assay. The dark green line in panels d and h depicts the cut-off at 6%, which is the threshold for seropositivity in the pseudoneutralising assay. Samples below 6% were assigned the value of 3. The points in red (negative) and blue (positive) in panels a, b, e, and f depict seropositivity based on the random forest algorithm using antibody levels to Nucleocapsid, Spike, and Recepter Binding Domain measured by the Luminex bead-based multiplex assay.

Fig. 3

Pairwise comparison of 2449 serum samples in three immunoassays. Grey highlighted numbers in panels a, d, and f provide an assessment of the agreement between the immunoassays. Numbers located at the top-right are samples classified as positive in both assays, and numbers bottom-left are samples classified as negative in both assays. Numbers in the top-left or bottom-right are discordant classified samples between two tests. The dark red dashed line in panels a, b and f is the cut-off defining seropositivity for the Luciferase-Linked ImmunoSorbent Assay. The dark green line in panels d, e, and f depicts the cut-off at 6%, which is the threshold for seropositivity based on the pseudoneutralising assay. Samples below 6% were assigned the value of 3. The points in red (negative) and blue (positive) depict seropositivity based on the bead-based multiplex assay.

Fig. 4

Epidemiological description of 2449 samples collected in North-eastern France, 2020. Panel a depicts the seroprevalence by age group and sex. In Panel b, the seroprevalence is stratified by month of sampling and in panel c by hospital.

Fig. 5

Antibody response to SARS-CoV-2 of 2449 samples collected in North-eastern France, 2020. Relative antibody units to four antigens shown by age measured with the LuLISA and the multiplex assay (Luminex). A random forests algorithm determined SARS-CoV-2 classification (positive in blue and negative in red). The smoothed line was fitted using the LOWESS method, with 95% confidence intervals denoted by the shaded region. The sixth graph shows the pseudo neutralisation activity by age.

Fig. 6

Relative antibody units to the four seasonal coronaviruses by age and seropositivity classification of 2449 samples collected in North-eastern France, 2020. The smoothed line was fitted using the LOWESS method, with 95% confidence intervals denoted by the shaded region.