Declining Levels of Neutralizing Antibodies Against SARS-CoV-2 in Convalescent COVID-19 Patients One Year Post Symptom Onset

Abstract

Major advances have been made in understanding the dynamics of humoral immunity briefly after the acute coronavirus disease 2019 (COVID-19). However, knowledge concerning long-term kinetics of antibody responses in convalescent patients is limited. During a one-year period post symptom onset, we longitudinally collected 162 samples from 76 patients and quantified IgM and IgG antibodies recognizing the nucleocapsid (N) protein or the receptor binding domain (RBD) of the spike protein (S). After one year, approximately 90% of recovered patients still had detectable SARS-CoV-2-specific IgG antibodies recognizing N and RBD-S. Intriguingly, neutralizing activity was only detectable in ~43% of patients. When neutralization tests against the E484K-mutated variant of concern (VOC) B.1.351 (initially identified in South Africa) were performed among patients who neutralize the original virus, the capacity to neutralize was even further diminished to 22.6% of donors. Despite declining N- and S-specific IgG titers, a considerable fraction of recovered patients had detectable neutralizing activity one year after infection. However, neutralizing capacities, in particular against an E484K-mutated VOC were only detectable in a minority of patients one year after symptomatic COVID-19. Our findings shed light on the kinetics of long-term immune responses after natural SARS-CoV-2 infection and argue for vaccinations of individuals who experienced a natural infection to protect against emerging VOC.

Keywords: COVID-19; SARS-CoV-2; antibody responses; humoral immunity; spike.

Figure 1

Sero-positivity of virus-specific IgG and IgM antibodies in 76 convalescent COVID-19 patients over time. The x-axis indicates the timeline following the onset of symptoms. Curves show the overall sero-positivity of IgG and IgM antibodies recognizing the RBD of the spike protein (anti-S IgG: green; anti-S IgM: purple) and the nucleoprotein (anti-N IgG: orange; anti-N IgM: blue). Numbers of samples (No.) at different time points are depicted under the diagram.

Figure 2

Longitudinal IgM and IgG antibody responses recognizing different SARS-CoV-2 proteins. IgM and IgG recognizing the RBD of the spike protein (S) or the nucleoprotein (N) were quantified by capture chemiluminescence immunoassays (CLIA) in 162 samples derived from 76 convalescent patients. (A–D) The plasma antibody levels of anti-S IgM, anti-N IgM, anti-S IgG, and anti-N IgG in convalescent patients. The numbers of samples at different time points were as follows: 31 (1-2 month), 11 (3-4 month), 26 (5-6 month), 9 (7-8 month), 9 (9-10 month), and 76 (11-12 month). (E–H) Repetitive sampling of 18 convalescent patients confirmed the kinetics of IgM and IgG responses recognizing SARS-CoV-2. The horizontal dotted line represents the cut-off value: 0.7 AU/mL (anti-IgM-S) and 1.0 AU/mL (anti-IgM-N, anti-IgG-S, and anti-IgG-N). P values were determined using a one-way ANOVA test. ns, no significance; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001.

Figure 3

Anti-S/N IgM and anti-S/N IgG antibody responses on year after disease onset in patients stratified according to age, sex, and disease severity. (A) Comparison of anti-S IgM, anti-N IgM, anti-S IgG, and anti-N IgG titers in patients stratified according to age; < 65 years (n=50, blue) and ≥ 65 years (n=26, yellow). (B) Comparison of anti-S IgM, anti-N IgM, anti-S IgG, and anti-N IgG titers in patients grouped according to sex; female (n=41, blue) and male (n=35, yellow). (C) Comparison of anti-S IgM, anti-N IgM, anti-S IgG, and anti-N IgG titers in patients stratified according to disease severity; non-severe (n=53, blue) and severe (n=23, yellow). Data are presented as median ± 95% CI. P values were determined applying a two-tailed Mann-Whitney U test. *p < 0.05.

Figure 4

Levels of neutralizing activity in convalescent COVID-19 patients. (A) The neutralizing activity was quantified in 137 samples obtained from 76 patients. Different colored boxes depict indicated SARS-CoV-2 neutralization titers at indicated times after symptom onset. Numbers of samples (No.) at different time points are shown below the diagram. (B) Neutralization activity one year post symptom onset in patients stratified according to age (upper: <65y, lower: ≥65y), sex (upper: male, lower: female), and disease severity (upper: non-severe, lower: severe). (C) Comparison of anti-S IgM, anti-N IgM, anti-S IgG, and anti-N IgG in the nAb-positive (n=31, yellow) and the nAb-negative population (n=42, blue) one year post symptom onset. Data are presented as median ± 95% CI. P values were determined with a two-tailed Mann-Whitney U test. *p < 0.05.

Figure 5

Neutralizing activity of 53 convalescent patients against the VOC B.1. 351. (A) The percentage of convalescent patients showing neutralizing activity against the VOC B.1.351. (B) Direct comparison of neutralization titers against the original SARS-CoV-2 strains (‘Wild type’) and VOC B.1.351 in 53 convalescent patients. P values were determined using a two-tailed Wilcoxon test. ns, no significance; ****p < 0.0001.

Figure 6

A schematic representation of SARS-CoV-2 antibody responses following symptomatic infection. In summary, humoral immune responses against SARS-CoV-2 decline over time in convalescent COVID-19 patients. While IgM antibody titers decline rapidly, SARS-CoV-2-specific titers of IgG showed a slower decay rate. Furthermore, minor neutralization against the VOC B.1.351 were detectable in convalescent COVID-19 patients compared to wild type SARS-CoV-2. The x-axis indicates the timeline following the onset of symptoms. The continuous curves depict the actual response (e.g., CLIA reactivity), while the dashed lines show the percentage of positivity for neutralizing antibodies (black), and anti-S IgM (purple), anti-N IgM (blue), anti-S IgG (green), and anti-N IgG (yellow line).