Wild-type SARS-CoV-2 neutralizing immunity decreases across variants and over time but correlates well with diagnostic testing

Abstract

Importance: The degree of immune protection against severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) variants provided by infection versus vaccination with wild-type virus remains unresolved, which could influence future vaccine strategies. The gold-standard for assessing immune protection is viral neutralization; however, few studies involve a large-scale analysis of viral neutralization against the Omicron variant by sera from individuals infected with wild-type virus.

Objectives: 1) To define the degree to which infection versus vaccination with wild-type SARS-CoV-2 induced neutralizing antibodies against Delta and Omicron variants.2) To determine whether clinically available data, such as infection/vaccination timing or antibody status, can predict variant neutralization.

Methods: We examined a longitudinal cohort of 653 subjects with sera collected three times at 3-to-6-month intervals from April 2020 to June 2021. Individuals were categorized according to SARS-CoV-2 infection and vaccination status. Spike and nucleocapsid antibodies were detected via ADVIA Centaur^® (Siemens) and Elecsys^® (Roche) assays, respectively. The Healgen Scientific^® lateral flow assay was used to detect IgG and IgM spike antibody responses. Pseudoviral neutralization assays were performed on all samples using human ACE2 receptor-expressing HEK-293T cells infected with SARS-CoV-2 spike protein pseudotyped lentiviral particles for wild-type (WT), B.1.617.2 (Delta), and B.1.1.529 (Omicron) variants.

Results: Vaccination after infection led to the highest neutralization titers at all timepoints for all variants. Neutralization was also more durable in the setting of prior infection versus vaccination alone. Spike antibody clinical testing effectively predicted neutralization for wild-type and Delta. However, nucleocapsid antibody presence was the best independent predictor of Omicron neutralization. Neutralization of Omicron was lower than neutralization of either wild-type or Delta virus across all groups and timepoints, with significant activity only present in patients that were first infected and later immunized.

Conclusions: Participants having both infection and vaccination with wild-type virus had the highest neutralizing antibody levels against all variants and had persistence of activity. Neutralization of WT and Delta virus correlated with spike antibody levels against wild-type and Delta variants, but Omicron neutralization was better correlated with evidence of prior infection. These data help explain why 'breakthrough' Omicron infections occurred in previously vaccinated individuals and suggest better protection is observed in those with both vaccination and previous infection. This study also supports the concept of future SARS-CoV-2 Omicron-specific vaccine boosters.

Keywords: COVID-19; SARS-CoV-2; antibody; nucleocapsid; spike; vaccine; variant of concern; viral neutralization.

Figure 1

Study Timeline and Emergence of known Variants of Concern for SARS-CoV-2. Timeline of visits related to vaccine availability and variant detection dates. VOC, variant of concern.

Figure 2

Viral neutralization across 3 timepoints for WT, Delta and Omicron SARS-CoV-2. (A-I) Viral neutralization across 3 timepoints for WT, Delta, and Omicron SARS-CoV-2. Each point represents the IC50 (log10 transformed) of an individual. Serum dilutions start at 1:20, visualized by the dotted line. Statistical analysis was performed using Kruskal-Wallis Multiple Comparisons. I, Infection; V, Vaccine; IC50, half maximal inhibitory concentration. (J). Results of a linear mixed effects model of log-transformed IC50 values analyzed according to infection, vaccination, or infection plus vaccination status. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Figure 3

Impact of Variants by Group. (A-F) Viral neutralization comparisons of WT, Delta, and Omicron SARS-CoV-2. Each point represents the IC50 (log10 transformed) of an individual. Serum dilutions start at 1:20, visualized by the dotted line. Results are grouped by variant (A-C) or by infection/vaccination status (D-F) for clarity. Statistical analysis was performed using Kruskal-Wallis Multiple Comparisons. I, Infection; V, Vaccine; IC50, half maximal inhibitory concentration. *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Figure 4

Impact of Time Since Infection or Vaccination. (A) Individuals at all timepoints with history of infection but without vaccination analyzed over time across all variants. (B) Individuals at all timepoints with history of vaccination but without prior infection analyzed over time across all variants. (C) Individual IC50s over time by variant for all participants with history of infection without vaccination. Line indicates the results of a linear mixed effects model of log-transformed IC50 values analyzing the role of time since infection controlling for vaccine status. (D) Individual IC50s over time by variant for all participants with history of vaccination without prior infection. Line indicates the results of a linear mixed effects model of log-transformed IC50 values analyzing the role of time since vaccination controlling for infection status. **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001.

Figure 5

Correlation Between Clinical Testing and Viral Neutralization. (A-F) IC50 results for each variant (WT, Delta, Omicron) categorized by spike antibody and nucleocapsid status (positive or negative) regardless of infection or vaccination status across all variants and time points. Serum dilutions start at 1:20, visualized by the dotted line. (G) Results of a linear mixed effects model of log-transformed IC50 values analyzing LFA results each for IgM and IgG, spike antibody status, and nucleocapsid antibody status all as dichotomous results.