Decreasing humoral response among healthcare workers up to 4 months after two doses of BNT162b2 vaccine.

Dear Editor,

We read with interest the articles by Harris Ross et al., and its comment by Capetti Amedeo et al., showing an antibody persistence up to 24 weeks and one year among COVID-19 infected individuals. In response to this journal, Tré-Hardy et al., and Salvagno Gian et al., presented slightly decreasing anti-S IgG decrease among vaccinated healthcare workers (HCWs) three months after mRNA vaccine first dose suggesting the need for potential new vaccine injection. Here we want to both provide new elements that strenghen this point with the analysis of vaccinated HCWs response up to five months after the first vaccine dose and add live virus neutralization data on several SARS-CoV-2 variants.

Thus, in the current study, we tested 138 sera samples from 17 BNT162b2 vaccinated HCWs up to 5 months after the first vaccine dose for anti-N and anti-S IgG (Abbott Diagnostics, Chicago, US) and pseudoneutralization activity (iFlash-2019-nCoV Nab assay, YHLO, Shenzen, China). Live virus neutralization assays with B, Alpha, Beta and Gamma SARS-CoV-2 strains were undertaken for a subset of 45 sera from 9 vaccinated HCWs. Decomplemented sera were subjected to serial two-fold dilution (1:25 to 1:12800), incubated with 50 μL of diluted virus (2 × 103 PFU/mL) in a 96-well plate at 37°C, 5% CO2 for 60 min. Then 3 × 104 cells Vero E6 cells (ATCC, reference R CRL-1586) were added and incubated (37°C, 5% CO2) until cytopathic effect assessment at day 4.

Among our vaccinated HCWs (median age 57 years old [IQR=24-52]), humoral response demonstrated a median anti-S titer at 728 AU/mL [343-1612] one month after the first dose and an antibody peak one month after the second dose at 11720 AU/mL [8350-20056] (Fig. 1 ). Then it gradually decreased, reaching a plateau 4 months after second dose (3059 AU/mL [2314-5124]). This decrease, five months after first vaccine dose, confirmed the trends observed by Tré-Hardy et al., and Salvagno Gian et al., at three months. The anti-S titers decrease was also confirmed by pseudoneutralization titers with a peak at 1593.1 AU/mL [788.4–1659.6], one month after the second dose followed by a plateau at 242.0 AU/mL [157.7-365.1]. When confirming those observations on live virus neutralization assay (Fig. 1). We also observed a pic of neutralizing anti-SARS-CoV-2 antibodies one month after the second vaccine dose (median neutralizing antibody titer of 1:12800 [1:3200–1:12800], 1:6400 [1:3200–1:6400] and 1:6400 [1:6400–1:12800] for B.1, Alpha and Gamma variants, respectively). However, we observed an even stronger decrease in neutralizing activity of sera, for all variants, leading to very low neutralizing titers at around 1:50 four months after the second vaccine dose for Alpha and Beta variants (1:50 [1:25–1:200] and 1:50 [1:25–1:100], respectively). A lower decline in neutralizing titers was also observed for ancestral strain and Gamma variant with median titers of 1:400 [1:200–1:800] and 1:100 [1:100–1:200], respectively. Regarding the Beta variant, neutralizing capacity of sera stayed significantly lower at 1 and 2 months after the second vaccine injection compared to ancestral SARS-CoV-2 strain (p=0.009 and p=0.014 respectively), Alpha variant (p=0.009 and p=0.034, respectively) and Gamma variant (p=0.009 and p=0.014, respectively). This difference of neutralization activity faded as neutralization titers dropped for all lineages 3 months after the second dose.

Fig. 1

Antibody kinetics and Neutralization titers among vaccinated HCWs up to 5 months after BNT162b2 vaccination initiation. Panel A depicts the anti-S IgG, panel B the Spike RBD-pseudoneutralization titers and panel C the live virus neutralization assay titers according to viral lineages.

To help routine antibody response follow-up, we evaluated the global concordance between anti-S IgG measurement, pseudoneutralization assay and live virus neutralization assay. The comparison between pseudo-neutralizing antibodies titers and anti-S titers was conducted on the 138 samples from vaccinated HCWs, 48 additional samples from COVID-19 patients and 16 pre-epidemic samples. We observed an overall agreement of 96% among vaccinated HCWs (n=133/138) and 100% among COVID-19 patients (n=48/48) and pre-pandemic samples (n=16/16) associated to a strong correlation between anti-S IgG and pseudo-neutralization titers with a Spearman coefficient Rho = 0.95 (p<0.0001) (Fig. 2 ). Regarding the 5 non concordant results in the HCWs group, all were anti-S IgG positive but negative for pseudoneutralization. They all were obtained less than 7 days after the first vaccine dose, compatible with an early low-RBD affinity antibody production. Interestingly, CPE neutralization assay demonstrated a strong neutralizing activity of vaccinated HCWs one month after the first vaccine dose against all variants (median IQR x4) despite relatively low anti-S (median IQR) and pseudoneutralization titers (median IQR) (Fig. 2). After the second vaccine dose, the neutralization titers were correlated with anti-S and pseudoneutralization titers (p<0.0001 for all lineages with rho's Spearman correlation test at 0.90, 0.86, 0.77 and 0.89 for B, Alpha, Beta and Gamma lineages respectively).

Fig. 2

Correlation between IgG anti-S with pseudoneutralization and live virus neutralization assays. Panel A depicts the correlation with pseudoneutralisation assay for vaccinated HCWs, depicted with black circles, COVID-19 diagnosed patients, depicted with dark gray squares, and prepandemic samples, depicted with light gray triangles. Panel B depicts the correlation with neutralization assay titers for tested SARS-CoV-2 lineages. Sampling timepoints after the vaccine first dose are indicated by the dots’ shapes and colors (D0: time of first vaccine dose, M1-2-3-5: one, two, three and five months after first vaccine dose).

In the current study on BNT162b2 vaccine among HCWs, a strong antibody response was observed two months after the first vaccine dose (i.e. one month after the second dose), decreasing drastically afterwards from two months up to four months after the first dose. From four to five months after the first dose, the antibody titers kept to slightly decrease while still presenting detectable anti-S IgG and pseudoneutralization positive results. If live-virus neutralization assays demonstrated the same kinetic, its results were more concerning as 0/9, 5/9, 6/9 and 2/9 HCWs presented no or low neutralization activities (i.e. with an effective dilution below 1:50) for the historical B, Alpha, Beta and Gamma strains, respectively.

These findings are of importance as another recent study on vaccinated HCWs in Israel demonstrated that reinfections were associated with lower antibody neutralizing titers and that lower neutralizing titers were also associated with higher viral load during infection that could lead to higher risk of transmission. Thus, the decrease in antibody and seroneutralization responses observed in the current work highlights the needs for additional vaccine doses evaluation. This is of peculiar importance for HCWs population which is at risk of transmission to fragile patients.

Funding

This study was supported in part by the ANRS|MIE (Agence Nationale de la Recherche sur le SIDA et les hépatites virales – Maladies Infectieuses Emergentes) and the Inserm UMR1137 unit.

CRediT authorship contribution statement

Valentine Marie Ferré: Conceptualization, Visualization, Investigation, Funding acquisition, Formal analysis, Data curation, Writing – original draft. Samuel Lebourgeois: Conceptualization, Visualization, Investigation, Funding acquisition, Formal analysis, Data curation, Writing – original draft. Reyene Menidjel: Investigation, Formal analysis, Data curation, Visualization. Gilles Collin: Investigation, Formal analysis, Data curation, Visualization. Houssem Redha Chenane: Investigation, Formal analysis, Data curation, Visualization. Manuella Onambele Guindi: Visualization, Investigation, Writing – review & editing. Yazdan Yazdanpanah: Visualization, Investigation, Writing – review & editing. Jean-François Timsit: Visualization, Investigation, Writing – review & editing. Charlotte Charpentier: Conceptualization, Visualization, Funding acquisition, Supervision. Diane Descamps: Conceptualization, Visualization, Funding acquisition, Supervision. Nadhira Fidouh: Conceptualization, Visualization, Investigation, Funding acquisition, Data curation, Supervision, Writing – original draft, Writing – review & editing. Benoit Visseaux: Conceptualization, Visualization, Investigation, Funding acquisition, Data curation, Supervision, Writing – original draft, Writing – review & editing.

Declaration of Competing Interest

The authors have no relevant competing interest to disclose in relation to this work.