The Impact of Prior Infection and Age on Antibody Persistence After Severe Acute Respiratory Syndrome Coronavirus 2 Messenger RNA Vaccine

Abstract

Determining the duration of immunity to severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines is critical for informing the timing of booster immunization. Many genetic and environmental factors could influence both the magnitude and persistence of the antibody response. Here, we showed that SARS-CoV-2 infection before vaccination and age affected the decay of antibody responses to the SARS-CoV-2 messenger RNA vaccine.

SARS, CoV, 2, COVID, 19, Antibody response, Vaccine

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) mRNA vaccines have been shown to be effective at preventing severe coronavirus disease 2019 (COVID-19) [1, 2]. The durability of vaccine-mediated protection or antibody immunity is not known, but antibody levels could be detected for the Moderna messenger RNA (mRNA) 1273 vaccine through 6 months [3]. Recent studies have demonstrated that individuals with prior SARS-CoV-2 infection had increased levels of SARS-CoV-2 spike protein binding and neutralizing antibodies after vaccination, compared with individuals with no previous infection [4–6]. In the current study, we determined antibody levels during SARS-CoV-2 vaccination and over 7 months after vaccination to identify factors that influence antibody decay rates.

METHODS

Antibody levels were determined after 2-dose vaccination with the Pfizer BNT162b2 SARS-CoV-2 mRNA vaccine in healthcare workers with laboratory-confirmed SARS-CoV-2 infection (by polymerase chain reaction) 30–60 days before vaccination (recent history of infection; n = 36) or without history of infection (no history of infection; n = 152) (Supplementary Table 1) [4, 7]. At baseline, serological testing was performed against the SARS-CoV-2 nucleocapsid protein to confirm history of infection and to identify potentially asymptomatic seropositive individuals, as described elsewhere [4, 7]. We collected peripheral blood at baseline (week 0), 3 weeks after the first dose (week 3), and 4 weeks after the second dose (week 7). In addition, 110 of the 188 individuals (25 with a recent history of infection and 85 with no history of infection) had blood samples collected later at weeks 16, 24, or 28.

The biospecimens from vaccine recipients were obtained at Children’s Mercy Kansas City, and their use was reviewed and approved by the Children’s Mercy institutional review board. Immunoglobulin G titers against SARS-CoV-2 spike proteins S1, S2, and receptor-binding domain (RBD) were determined from plasma using a multiplex bead-binding assay (Milliplex SARS-CoV-2 Antigen Panel 1 IgG, Millipore) [4, 7, 8]. As a proxy for measuring virus-neutralizing antibodies, we used an in vitro assay that allows indirect detection of potential SARS-CoV-2–neutralizing antibody through determination of antibody blocking of the SARS-CoV-2 RBD binding to the host receptor angiotensin-converting enzyme 2 (SARS-CoV-2 Surrogate Virus Neutralization test kit; Genscript) [4, 7–9]. Antibody half-life was calculated using an exponential decay model computed using the lme4 package in R software, which assumes a steady decay rate over time [3] (Supplementary Methods).

RESULTS

Individuals with recent SARS-CoV-2 infection before vaccination had significantly higher levels of binding antibody to the SARS-CoV-2 spike proteins S1, S2, and RBD at baseline and after primary immunization (week 3), compared with individuals with no history of infection (Figure 1A). After the second vaccine dose, the levels still differed significantly between the groups for S1 and S2 but not for RBD. However, the magnitude of difference between the two groups at week 7 was small, indicating that a second dose was indeed necessary in individuals without prior infection to achieve binding antibody levels equivalent to those with recent history of infection (Figure 1A and Supplementary Table 2). At week 28, both groups had decreased levels of antibody to the S1, S2, and RBD proteins, but levels remained significantly higher in those with a recent history of infection before vaccination than in those without prior infection (Figure 1A and Supplementary Table 2). Next, we measured viral host receptor blocking as a proxy for neutralizing antibodies. We observed similar dynamics of potential neutralizing antibodies, with higher primary response titers at week 3 and higher levels at weeks 24 and 28 in individuals with recent infection before vaccination (Figure 1B and Supplementary Table 3).

Figure 1.

Antibody levels after severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) messenger RNA (mRNA) vaccination. A, Multiplex bead-based antibody binding assay that measures the immunoglobulin G (IgG) antibody response to the SARS-CoV-2 spike subunit 1 (S1), spike subunit 2 (S2), and receptor-binding domain (RBD). Weeks of the study are displayed along the x-axis, starting with the baseline at week 0. Arrows indicate the timing of immunization with coronavirus disease 2019 (COVID-19) mRNA vaccine. Individual values of median fluorescence intensity (MFI) are calculated; background subtraction has been used to remove nonspecific signal. Geometric mean titers are displayed for each group (recent history of SARS-CoV-2 infection before vaccination, red; no history of infection before vaccination, blue). Error bars represent 95% confidence intervals (CIs). *P < .05 (Wilcoxon-Mann-Whitney test corrected for multiple comparisons [false discovery rate; Benjamini correction]) for comparison of groups at each time point. B–D, Neutralization antibody proxy assay that determines the level of antibodies that block binding of the spike protein RBD to the human host receptor angiotensin-converting enzyme 2 (ACE2), expressed as the percentage of binding that was blocked relative to a control with no plasma (representing maximum binding). The assay threshold for positivity was 30%, indicating the presence of neutralizing antibodies. Arrows indicate the timing of immunization with COVID-19 mRNA vaccine. Geometric mean titers of the percentage of blocking are displayed for each group. B, Individuals with recent history of SARS-CoV-2 infection before vaccination, red; no history of infection before vaccination, blue. C, Individuals with no history of infection before vaccination stratified by sex. D, Individuals with no history of infection before vaccination stratified by age; error bars represent 95% CIs. *P < .05 (Wilcoxon-Mann-Whitney test corrected for multiple comparisons [false discovery rate; Benjamini correction]) for comparison of groups immune at each time point. E, Rate of decay and half-life of antibodies from week 7 to week 28 for each group and immune assay using an exponential decay model, with antibody half-life displayed in weeks.

In individuals with no history of infection before vaccination, we did not find significant differences in antibody magnitude between men and women at any time point measured (Figure 1C). However, when we stratified individuals into two age groups, we found that individuals who were 18–49 years old had significantly higher blocking antibody titers at week 3 and week 28 than older individuals who were ≥50 years old (Figure 1D).

Antibody half-life after vaccination was calculated using an exponential decay model and varied in duration depending on the binding or blocking assay used (Figure 1E). In all cases, we observed longer antibody half-life in individuals with prior COVID-19 before vaccination than in individuals with no infection history; and in individuals aged 18–49 years compared with older individuals (Figure 1E and Supplementary Tables 4 and 5). The lowest antibody half-life calculated was 24.76 weeks for individuals with a recent history of infection and 13.60 weeks for those with no history of infection using the binding measures over time to the S1 spike subunit (Figure 1E and Supplementary Tables 4 and 5).

Discussion

These data demonstrated that individuals who had SARS-CoV-2 infection before vaccination or younger individuals had significantly higher levels of antibodies after primary immunization with a SARS-CoV-2 mRNA vaccine and had significantly longer antibody half-lives measured at 7 months after vaccination. The rate of antibody decay observed here are consistent with reports of other vaccine platforms and convalescent individuals after infection [3, 10]. In the current study, individuals had recent infection within 60 days before the administration of vaccine. It will be critical to determine whether individuals with SARS-CoV-2 infection >60 days before vaccination also have higher antibody responses that are maintained longer. The serological assays used to measure binding antibody levels and surrogate neutralizing antibodies were performed using SARS-CoV-2 spike proteins with the original amino acid sequences and do not capture differences in antibody responses to any of the emerging viral variants.

Future studies determining the cross-reactivity of antibody levels to viral variants will be required to understand the impact of potential antibody escape mutants. Antibody levels that correlated with vaccine efficacy or the breadth of response to viral variants have not been determined for SARS-CoV-2 infection or vaccines. These data suggest that the duration of protective antibody immunity may be influenced by prior infection history, age, and other factors that may affect timing of additional immunization. Thus, ongoing studies monitoring immune responses will be required to determine optimal booster vaccine strategies.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Acknowledgments. The authors thank all the healthcare workers that participated in this study. Special thanks to Occupational Health and Children’s Mercy Research Institute for their support of this study. Daniel Louiselle, MSc, Nick Nolte, BSc, Rebecca Biswell, BSc, assisted with peripheral blood sample processing and storage. Bradley Beldon, BSc, Angela Myers, MPH, MD, and Jennifer Schuster, MD, were involved in overseeing study design and clinical implementation with the mRNA vaccine. The vaccinee biospecimens were collected under a clinical study at Children’s Mercy Kansas City and reviewed and approved by the Children’s Mercy Institutional Review Board (nos. 00001670 and 00001317).

Disclaimer. The funding body had no role in collecting, analyzing, interpreting, or writing the manuscript.

Financial support. This work was supported by internal institutional funds from Genomic Medicine Center, Children’s Mercy Research Institute, and Children’s Mercy Kansas City. E. G. holds the Roberta D. Harding & William F. Bradley, Jr, Endowed Chair in Genomic Research.

Potential conflicts of interest. All authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest.

Author notes

R. S. and T. B. contributed equally to this work.