Durability of Immune Response After COVID-19 Booster Vaccination and Association With COVID-19 Omicron Infection

Abstract

Importance: The BNT162b2 two-dose vaccine (BioNTech/Pfizer) has high effectiveness that wanes within several months. The third dose is effective in mounting a significant immune response, but its durability is unknown.

Objective: To compare antibody waning after second and third doses and estimate the association of antibody kinetics with susceptibility to infection with the Omicron variant of SARS-CoV-2.

Design, setting, and participants: In a prospective longitudinal cohort study in a tertiary medical center in Israel, health care workers who received the BNT162b2 vaccine were followed up monthly for IgG and neutralizing antibody levels. Linear mixed models were used to compare antibody titer waning of second and third doses and to assess whether antibody dynamics were associated with Omicron transmission. Avidity, T cell activation, and microneutralization of sera against different variants of concern were assessed for a subgroup.

Exposure: Vaccination with a booster dose of the BNT162b2 vaccine.

Main outcomes and measures: The primary outcome was the rate of antibody titer change over time, and the secondary outcome was SARS-CoV-2 Omicron variant infection, as confirmed by reverse transcriptase-polymerase chain reaction.

Results: Overall, 4868 health care workers (mean [SD] age, 46.9 [13.7] years; 3558 [73.1%] women) and 3972 health care workers (mean [SD] age, 48.5 [14.1] years; 996 [74.9%] women) were followed up for 5 months after their second and third vaccine doses, respectively. Waning of IgG levels was slower after the third compared with the second dose (1.32%/d [95% CI, 1,29%/d to 1.36%/d] vs 2.26% [95% CI, 2.13%/d 2.38%/d]), as was waning of neutralizing antibody levels (1.32%/d [95% CI, 1.21%/d to 1.43%/d] vs 3.34%/d [95% CI, 3.11%/d to 3.58%/d]). Among 2865 health care workers assessed for Omicron incidence during an additional 2 months of follow-up, lower IgG peak (ratio of means 0.86 [95% CI, 0.80-0.91]) was associated with Omicron infection, and among participants aged 65 years and older, faster waning of IgG and neutralizing antibodies (ratio of mean rates, 1.40; [95% CI, 1.13-1.68] and 3.58 [95% CI, 1.92-6.67], respectively) were associated with Omicron infection. No waning in IgG avidity was observed 112 days after the third dose. Live neutralization of Omicron was lower compared with previous strains, with a geometric mean titer at the peak of 111 (95% CI, 75-166), compared with 942 (95% CI, 585-1518) for WT, 410 (95% CI, 266-634) for Delta; it demonstrated similar waning to 26 (95% CI, 16-42) within 4 months. Among 77 participants tested for T cell activity, mean (SD) T cell activity decreased from 98 (5.4) T cells/106 peripheral blood mononuclear cells to 59 (9.3) T cells/106 peripheral blood mononuclear cells.

Conclusions and relevance: This study found that the third vaccine dose was associated with greater durability than the second dose; however, Omicron was associated with greater resistance to neutralization than wild type and Delta variants of concern. Humoral response dynamics were associated with susceptibility to Omicron infection.

Figure 1.. Recruitment of Participants, Testing, and Follow-up

This study involved a prospective cohort of health care workers (HCWs) who received the third dose of the BNT162b2 vaccine and underwent at least 1 serologic assay afterward. A, The schedule for serological follow-up is presented. B, Serological follow-up of participants eventually infected with the Omicron variant is presented. Patients who received a fourth dose of vaccine and those who had an early breakthrough infection were censored. Serological follow-up of individuals who were infected that was obtained after infection was censored as well. IgG indicates IgG antibodies; SMC, Sheba Medical Center; VOC, variant of concern.

Figure 2.. Humoral Response After Second and Third Doses

A, The distribution of observed IgG antibodies after the second and third dose (points) and the expected geometric mean titer (GMT) as estimated by a model adjusted by age (lines) are presented. BAU indicates binding antibody unit; RBD, receptor binding domain. B, The distribution of observed neutralizing antibodies after the second and third dose (points) and the expected GMT as estimated by a model adjusted by age (lines) are presented. C, Antibody avidity 7 to 28 days after the second dose, 7 to 28 days after the third dose, and 85 to 112 days after the third dose are presented. Dots indicate observed results; horizontal black lines, means; whiskers, 95% CIs. D, The microneutralization assays against wild type (WT), Delta, and Omicron variants are presented, consisting of microneutralization of sera of 25 participants against WT, Delta, and Omicron variants at 1, 2, 3, and 4 months after the third dose of vaccine. Dashed horizontal line indicates cutoff for diagnostic positivity; dots, observed results; horizontal black lines, geometric means; whiskers, 95% CIs.

Figure 3.. Distribution of Antibodies 150 Days After Third Doses Among Individuals Infected and Not Infected

A, The distribution of observed IgG antibodies after the third dose among participants eventually infected and those who remained uninfected (points) and the expected geometric mean titer as estimated by a model adjusted by age are presented. B, The distribution of observed neutralizing antibodies after the third dose among participants eventually infected and those who remained uninfected (points) and the expected geometric mean titer as estimated by a model adjusted by age. BAU indicates binding antibody unit; RBD, receptor binding domain.