SARS-CoV-2 BNT162b2 vaccine-induced humoral response and reactogenicity in individuals with prior COVID-19 disease

Abstract

BACKGROUNDMost individuals with prior COVID-19 disease manifest long-term protective immune responses against reinfection. Accordingly, we tested the hypothesis that humoral immune and reactogenicity responses to a SARS-CoV-2 mRNA vaccine differ in individuals with and without prior COVID-19 disease.METHODSHealth care workers (n = 61) with (n = 30) and without (n = 31) prior COVID-19 disease received two 30 μg doses of Pfizer BNT162b2 vaccine 3 weeks apart. Serum IgG antibody against the spike receptor-binding domain; serum neutralizing activity; and vaccine reactogenicity were assessed longitudinally every 2 weeks for 56 days after the first injection.RESULTSThe COVID-19 group manifested more rapid increases in spike IgG antibody and serum neutralizing activity after the first vaccine dose but showed little or no increase after the second dose compared with the infection-naive group. In fact, spike IgG was at its maximum level after the first dose in 36% of the COVID-19 group versus 0% of the infection-naive group. Peak IgG antibody levels were lower but appeared to fall more slowly in the COVID-19 group versus the infection-naive group. Finally, adverse systemic reactions, e.g., fever, headache, and malaise, were more frequent and lasted longer after both the first and second injection in the COVID-19 group than in the infection-naive group.CONCLUSIONIndividuals with prior COVID-19 disease demonstrate a robust, accelerated humoral immune response to the first dose but an attenuated response to the second dose of BNT162b2 vaccine compared with controls. The COVID-19 group also experienced greater reactogenicity. Humoral responses and reactogenicity to BNT162b2 differ qualitatively and quantitatively in individuals with prior COVID-19 disease compared with infection-naive individuals.FUNDINGThis work was supported by Temple University institutional funds.

Keywords: Adaptive immunity; COVID-19.

Figure 1. Spike RBD IgG antibody responses to the BNT162b2 vaccine in COVID-19 and control groups.

(A) Group mean ± 1 SEM responses. Vertical arrow indicates time of second vaccine injection. Note that the time course of spike RBD IgG antibody response to vaccination was significantly different between COVID-19 and control groups (P < 0.0001; linear mixed-effects model for repeated measures). Differences in spike IgG antibody levels were significant prevaccine (P < 0.0003) and at day 14 (**P < 0.0002). However, spike IgG levels were similar at days 28, 42, and 56 (P > 0.10 for all 3 comparisons). Sample size in the COVID-19 group was as follows: day 14 (n = 25), day 28 (n = 28), day 42 (n = 28); and day 56 (n = 28). Sample size in the control group was as follows: day 14 (n = 25), day 28 (n = 28), day 42 (n = 30), and day 56 (n = 31). (B) Spike RBD IgG antibody responses to vaccine in individuals with prior COVID-19 disease and control individuals. Note the considerable interindividual variability in both groups.

Figure 2. Relationship between vaccine-induced peak spike RBD IgG antibody and interval after onset SARS-CoV-2 symptoms in individuals with COVID-19.

There was no discernible relationship (r² = 0.01 by linear regression). Of note, 2 of the 29 individuals in the COVID-19 group were asymptomatic. Hence, no after onset of SARS-CoV-2 symptoms (POS) value is available.

Figure 3. Serum neutralizing activity in COVID-19 and control groups.

Pseudovirus uptake in HEK293 ACE2-overexpressing cells was assessed from luciferase activity, i.e., relative light units (RLU), and is shown on the y axis. Serum dilution is shown on the x axis. The 100% control value on the y axis represents maximal virus uptake occurring in the absence of serum. IC₅₀ was calculated using sigmoidal 4-factor polynomial, nonlinear regression. Data are shown prevaccination (A) and day 14 (B) and day 42 (C) after first vaccine dose. Prevaccine, neutralizing activity in the COVID-19 group (n = 7) tended to be greater than that in the control group (n = 7) (IC₅₀ 7 × 10^–3 vs. 2 × 10^–2 dilution, respectively) but was not statistically significantly different (P = 0.20 by 2-way ANOVA). At day 14 after first injection, neutralizing activity increased greatly in the COVID-19 group (n = 21) but was unchanged in the control group (n = 21) (IC₅₀ = 3.6 × 10^–4 vs. 1 × 10^–2 dilution, respectively; P < 0.03 by 2-way ANOVA for comparison of the 2 groups). In contrast, at day 42, neutralizing activity increased greatly in the control group (n = 21) but only slightly in the COVID-19 group (n = 21) (IC₅₀ = 5 × 10^–4 vs. 3 × 10^–4 dilution, respectively) and was again not statistically significantly different (P = 0.11 for comparison of the 2 groups by 2-way ANOVA). For days 14 and 42, the same 21 individuals with prior COVID-19 disease and 21 control individuals were studied, and for both groups each data point is the mean ± SEM for 3 pools of 7 individuals each.

Figure 4. Reactions to the first and second Pfizer BNT162b2 mRNA vaccine in COVID-19 and control groups.

(A) Prevalence and severity of systemic and local reactions. (B) Duration of systemic symptoms (mean ± 1 SEM for each group). (C) Frequency of medication use. Dark color (red or blue) indicates reaction scores of more than 6 severity; light colors indicate scores of equal to or less than 5. Brackets indicate statistical comparisons across groups assessed by Fisher’s exact test. Lines indicate comparisons within groups assessed by McNemar’s test for paired comparisons. *P < 0.05; **P < 0.01; †P < 0.001. For all comparisons, the COVID-19 group, n = 30; control group, n = 31.