Elicitation of neutralizing antibodies and IgG4 subclass switching following booster vaccination with ancestral COVID-19 mRNA vaccines does not reduce breakthrough infections

Abstract

Vaccination with COVID-19 mRNA vaccines generates robust antibody responses, but the impact of prior infection on the quality of these responses, particularly immunoglobulin G (IgG) subclass profiles remain unclear. A longitudinal study was conducted to compare humoral immune responses in SARS-CoV-2 infection-naïve and pre-immune (previously infected) individuals following a two-dose mRNA vaccine primary series and a booster dose. Anti-spike receptor-binding domain (RBD) IgG levels, neutralizing antibody titers against the vaccine-matched (Wuhan-Hu-1) virus and the Omicron BA.1 variant, and IgG1-IgG4 subclass distributions over time were measured. After two doses, pre-immune participants had higher anti-RBD IgG (2-fold, p < .001) and neutralization titers than naïve participants (GMT±SD; 6407 ± 4 vs 5706 ± 3.5). Notably, 89.7% of pre-immune versus 28.1% of naïve sera neutralized Omicron BA.1 (p < .0001). However, a third (booster) raised antibody levels in naïve participants to a statistically similar titer to pre-immune participants (p > .05). The booster vaccination also markedly enhanced the titers of cross-neutralizing antibodies against Omicron BA.1 in both vaccine groups. This increased the proportion of responders from 31.8% to 95.5% in naïve participants. Booster vaccinations induced significant IgG4 titers in naïve (p < .0001), but not in pre-immune participants (p > .05) compared to primary series of vaccination levels. Both vaccinated naïve participants and pre-immune vaccinated participants had a breakthrough infection within one year following the booster vaccination (36.4% vs 25%, p > .05 respectively). We report that the presence of IgG4 antibodies, specifically in naïve individuals, did not alter in vitro virus neutralizing response against ancestral WH-1 and Omicron BA.1 variant, with comparable breakthrough infection rates.

Keywords: COVID-19; IgG subclass switching; SARS-CoV-2; mRNA vaccine; naïve; pre-immune; virus neutralization.

Figure 1.

(A) schematic representation of the vaccination schedule and sera collection time points as shown (created with Biorender.com). (B) Log10 transformed WH-1 RBD-binding IgG (µg/mL) levels before and after each vaccination in naïve and pre-immune participants at different time points. Fold changes (x) in IgG levels indicated pre-immune IgG level compared to naïve IgG levels (pre-imm/naïve) at corresponding time points. A parametric statistical analysis performed with two-way ANOVA test to compare differences between naïve (n = 32) and pre-immune (n = 39) groups at time points indicated. Statistical significance is denoted by asterisks (**p < .01, ***p < .001, ****p < .0001), and non-significant (ns) differences are labeled accordingly. Error bars represent the geometric mean titer (GMT) with standard deviation (SD). The horizontal dotted lines indicating the threshold of the assay (1.137 µg/mL) while vertical dotted line is representing the vaccination time points.

Figure 2.

Wuhan-Hu-1 (WH-1) pseudovirus neutralization (PsVN) responses in naïve and pre-immune individuals following vaccination. Longitudinal Log10-transformed PsVN in (A) naïve and (B) pre-immune individuals shown and compared before and after completion of primary series of vaccination (first and second) and booster vaccination (third). (C) comparison of neutralization responses between naïve and pre-immune individuals at different time points. PsVN titers normalized to 20 sera dilution if they were less than 20. The horizontal dotted lines indicate the assay threshold (20 IC50 before Log10-transformed). For the statistical analysis data shown in PsVN IC50 (A-B), a paired ANOVA model performed. For the data in figures C, un-paired ANOVA model performed to assess differences between the naïve and pre-immune groups. Statistical significance is denoted by asterisks (**p < .01, ****p < .0001), and non-significant (ns) differences are labeled accordingly. The numbers of participants tested were as follows: for naïve participants, before the first and after the second doses, n = 32; and before and after the third dose, n = 22. For pre-immune participants, before the first and after the second doses, n = 39, and before and after third dose, n = 16.

Figure 3.

Comparison of pseudovirus neutralization (PsVN) IC50 and cross-reactivity between Wuhan (WH-1) and Omicron BA.1 variant. (A) Log10-transformed PsVN IC50 titers (log10) against WH-1 and Omicron BA.1 variant in naïve and pre-immune participants shown after second, (B) before the third vaccine, and (C) after the third vaccine dose. Pie charts illustrate the percentage of participants who exhibited detectable (>20 IC50) neutralization against the specified SARS-CoV-2 pseudovirus. Fold changes (x) in neutralization shown on the figures, were calculated by comparing geometric mean of titers (GMT). Data are represented as GMT with standard deviation (SD). Log-transformed PsVN IC50 titers (log10) were compared between WH-1 and BA.1 using one-way ANOVA with significance levels shown: ***p < .001, ****p < .0001, and ns: not significant.

Figure 4.

IgG subclass responses and comparison of SARS-CoV-2 infection rates between naïve and pre-immune individuals. Serum levels of (A) IgG1, (B) IgG2, (C) IgG3, and (D) IgG4 subclasses were represented with O.D. against the WH-1 spike RBD following the second (post-2nd) and third (post-3rd) vaccine doses. The assay cutoff (dotted line) for each IgG subclass was determined using negative serum samples from individuals who tested seronegative for anti-RBD antibodies. Significance was determined using a paired one-way ANOVA with; ns = not significant, ****p < .0001. (E) comparison of SARS-CoV-2 infection rates between naïve and pre-immune individuals following booster vaccination. Bar graph shows the proportion of individuals who experienced SARS-CoV-2 infection versus those who remained uninfected (protected) during the one-year follow-up period post-booster. Infection was confirmed by nucleic acid amplification test (NAAT) positivity. Fisher’s exact test was performed to compare infection versus non-infection rates between naïve and pre-immune individuals (p > .05; ns: not significant).