Serum neutralization of SARS-CoV-2 Omicron sublineages BA.1 and BA.2, and cellular immune responses 3 months after booster vaccination.

OBJECTIVES: We investigated serum neutralizing activity against BA.1 and BA.2 Omicron sublineages, and T cell response before and 3 months after booster vaccine in healthcare workers (HCWs).
METHODS: HCWs aged 18-65 years, vaccinated and boosted with the BNT162b2 vaccine were included. Anti-SARS-CoV-2 IgG levels and cellular response (through IFNγ ELISpot assay) were evaluated in all participants, and neutralizing antibodies against Delta, BA.1 and BA.2 were evaluated in participants with at least one follow-up visit 1 or 3 months after the booster dose.
RESULTS: Among the 118 HCWs who received the booster dose, 102 and 84 participants attended the 1-month and 3-month visits, respectively. Before the booster vaccine dose, a low serum neutralizing activity against Delta, BA.1, and BA.2 was detectable in only n=39/102 (38.2%), n=8/102 (7.8%) and n=12/102 (11.8%) of the participants, respectively. At 3 months, neutralizing antibodies were detected in n=84/84 (100%), n=79/84 (94%) and n=77/84 (92%) against Delta, BA.1 and BA.2, respectively. Geometric mean titers (GMTs) of neutralizing antibodies against BA.1 and BA.2 were 2.2- and 2.8-fold reduced compared to Delta. From 1 to 3 months after the booster dose, participants with a recent history of SARS-CoV-2 infection (n=21/84) had persistent levels of S1 reactive specific T cells and of neutralizing antibodies against Delta and BA.2, and 2.2-fold increase in neutralizing antibodies against BA.1 (p= 0.014). Conversely, neutralizing antibody titers declined from 1 to 3 months after booster dose in individuals without any recent infection, against Delta (2.5-fold decrease, p<0.0001), BA.1 (1.5-fold, p=0.02) and BA.2 (2-fold, p<0.0001).
CONCLUSIONS: The booster vaccine dose provided significant and similar response against BA.1 and BA.2 Omicron sublineages, but the immune response declines in the absence of recent infection.

Introduction

The Omicron (B.1.1.529) SARS-CoV-2 variant has emerged in late 2021, and has rapidly outcompeted the already highly transmissible Delta variant.

The first sublineage of this variant, BA.1, became dominant in Europe between December 2021 and January 2022 and has generated serious concern about the efficacy of vaccines, due to substantial escape from neutralizing antibodies. Indeed, Omicron displayed numerous epitopic modifications of the spike protein, and as available COVID-19 vaccines were prepared with the original lineage, neutralizing activity against Omicron was absent or very low even in vaccinated people [1,2].

The subsequent sublineage BA.2, has rapidly replaced the previous one in Europe around March-April 2022, and rates of new sublineages BA.4 and BA.5 are increasing by June 2022 [3]. BA.2 has shown a selective advantage over BA.1, especially an enhanced transmissibility, and can reinfect previously BA.1-infected individuals [4,5].

Recent data have shown that beyond 6 months after prime vaccination, neutralizing antibodies against Omicron sublineages were undetectable or at very low levels; however, neutralizing antibody titers increased significantly few weeks after booster vaccination (third dose) [[6], [7], [8]].

In this study, we evaluated neutralizing antibody responses against the BA.1 and BA.2 sublineages of the Omicron variant, in parallel with the previous Delta variant in a cohort of healthcare workers (HCWs) who had been vaccinated and boosted with the BNT162b2 mRNA vaccine (Pfizer–BioNTech). In addition, cellular responses to SARS-CoV-2 were investigated, and the occurrence of infection after booster dose analyzed.

Methods

This study is part of the MONITOCOV project in which the protocol amendments concerning the present data, were approved by the Ile-De-France V (ID‐CRB 2021-A00119-32) ethics committee.

Health care workers were consecutively included in the study if they had no significant chronic disease and no medication which could influence immune responses (including steroids, immunosuppressive therapies, ..), and were aged 18–65 years. All participants initially received the two-dose BNT162b2 vaccination at a 3-weeks dosing interval. The booster dose was administered to participants in accordance with French national guidelines (at least six months after primary vaccination course). Data from samples collected before and 3 months after primary vaccination were reported previously [9], and data from samples collected before booster, and at 1- and 3-month post-booster are reported in this study.

Anti-SARS-CoV-2 IgG levels, as well as neutralizing antibodies (using a live virus neutralization assay) against Delta variant, and BA.1 and BA.2 Omicron sublineages were evaluated. Cellular responses were evaluated through IFNγ ELISpot assay.

Detailed methods are provided in the supplementary material.

Results

A total of 118 individuals who received the booster dose were included. Samples were collected at 1- month after booster for 102 subjects, and only 84 participants attended the 3-month visit.

The median (interquartile range, IQR) time between the initial vaccination (second dose) and the booster dose was 254 (248-265) days.

Three months after the primary vaccination course (D1/D2), the median level (IQR) of anti-SARS-CoV-2 IgG was 1125 (653;1828) BAU/mL. This value declined to 290 (200;534) BAU/mL before the third dose, and then peaked to more than 2080 BAU/mL, 1 and 3 months after the booster dose (Figure 1 A). The counts of S1 reactive T cells also increased 1 and 3 months after the booster dose (Figure 1A).

Fig 1

Specific antibody and T cell response 1-month and 3-month after booster BNT162b2 mRNA vaccine. (A) Anti-S1 IgG levels (blue, left Y axis) and S1 specific reactive T cells (black, right Y axis), from before initial vaccination (Pre D1/D2) to 3-month post booster vaccine dose. (B) Neutralization of the Delta, BA.1 Omicron and BA.2 Omicron variants before and 1-month and 3-month after booster BNT162b2 mRNA vaccine dose. Positive threshold (titer ≥ 20) is shown with dotted line. Individual values, geometric mean neutralization titers (GMTs) and 95% confidence intervals are shown. Wilcoxon signed rank test was used for within-subject comparisons. **p-values <0.01; ***p-values <0.001; ****p-values <0.0001; ns, not significant. (C) S1 specific reactive T cell counts before, 1-month and 3-month after booster BNT162b2 mRNA vaccine. Individual values and median (interquartile range)] are shown. N (left panel) and S1 (right panel) specific reactive T cells in healthcare workers without (circle) or with (square) COVID-19 after the booster dose. Mann-Whitney U test was performed to compare T cell counts between both groups. **p=0.006; ****p-values < 0∙0001. (D) Neutralization of the Delta, BA.1 Omicron and BA.2 Omicron variants in healthcare workers without (circle) or with (square) COVID-19 after the booster dose. Positive threshold (titer ≥ 20) is shown with dotted line. Geometric mean neutralization titers (GMTs) and 95% confidence intervals are shown. A mixed model ANOVA was performed to compare both groups. ****p-values <0.0001.

As shown in Figure 1B, before the booster vaccine dose, serum neutralizing activity against Delta, BA.1, and BA.2 was detectable in only n=39/102 (38.2%), n=8/102 (7.8%) and n=12/102 (11.8%) of the participants, respectively, with low neutralizing antibody titers ranging from 20 to 160. Geometric mean titers (GMTs) were below the limit of detection of 20 for all variants.

One month after the booster dose, a neutralizing activity against Delta, BA.1 and BA.2 was observed in n=102/102 (100%), n=99/102 (97%) and n=101/102 (99%) of tested sera, respectively. Also, GMTs of neutralizing antibodies (95% confidence interval, CI95%) significantly increased to 465 (373;579) for Delta, 105 (85;131) for BA.1 and 99 (82;121) for BA.2, indicating that the GMTs against Delta were 4.4- and 4.7-fold higher compared to BA.1 (p < 0.0001) and BA.2 (p < 0.0001), respectively.

Three months after the booster dose, neutralizing antibodies were detected in 100% (n=84/84), 94% (n=79/84) and 92% (n=77/84) against Delta, BA.1 and BA.2, respectively, and the GMTs were 260 (203;334), 102 (76;136) and 69 (54;88) respectively. GMTs against BA.1 and BA.2 were 2.6- and 3.8-fold reduced compared to Delta 3 months after the booster dose (p < 0.0001 in both comparisons).

Within 3 months following the booster dose, a SARS-CoV-2 infection was confirmed or strongly suspected in 21 subjects (presumably Omicron infections as this variant represented > 95% of cases during this period). The infection was confirmed by a positive RT-PCR in 10 individuals, with the infection occurring during the first 4 weeks (1 subject) or during the next 8 weeks (9 subjects) following the booster vaccination. An asymptomatic infection, not confirmed by RT-PCR for this reason [10], and not detectable by serological tests using antigens from the Spike protein, was strongly suspected in the remaining 11 individuals by a significant increase of N reactive specific T cells in IFNγ ELISpot assay, as the N protein is not an antigen targeted by BNT162b2 (Figure 1C).

The subgroup analysis showed that 3-month post-booster dose, S1 reactive specific T cells (Figure 1C) and neutralizing antibody levels against all variants were higher in the group with a history of post-booster infection compared to participants without recent infection (Figure 1D, Supplementary Table S1). Indeed, GMT (CI95%) of neutralizing titers declined from 1 to 3 months after booster dose in individuals without any recent infection for Delta (482 (366; 635) vs 190 (143;252), 2.5-fold decrease, p<0.0001), for BA.1 (110 (85;142) vs 74 (85;142), 1.5-fold decrease, p=0.02) and BA.2 (104 (81;133) vs 52 (39;69), 2-fold decrease, p<0.0001). Conversely, participants with a recent history of SARS-CoV-2 infection had persistent levels of S1 reactive specific T cells and of neutralizing antibodies against Delta and BA.2, and neutralizing antibodies against BA.1 increased from 128.5 (72; 229) to 277 (137; 557) (2.2-fold increase, p=0.014). (Figure 1D, Supplementary Table S1).

Discussion

In this work, we have evaluated the vaccine-induced response against Omicron sub-variants. We found that 8.5 months (median) after primary vaccination with two-doses of an mRNA vaccine, HCWs showed minimal neutralization antibody response to Delta variant as well as BA.1 and BA.2 Omicron sublineages. Following the booster dose, the participants exhibited a stronger neutralizing capacity not only against Delta variant but also BA.1 and BA.2. This is consistent with prior reports regarding BA.1 [2,[11], [12], [13], [14], [15]] and BA.2 [[6], [7], [8]]. However, the median neutralizing antibody titers against BA.1 and BA.2 at 1- and 3-month post-booster were significantly reduced, as compared to the previous Delta variant, indicating a reduced effectiveness of booster vaccination against Omicron variant.

Interestingly, at 1-month post-boost, we observed similar neutralizing antibody levels against BA.1 and BA.2, highlighting that the immune response induced by the BNT162b2 vaccine was similar for both variants. This finding of the early weeks post-booster is in agreement with other reports [6,7]. However, the decline over time seemed to be more pronounced for neutralizing activity against BA.2. This observation which needs to be confirmed could partially explain the surge of BA.2 and the replacement of BA.1 which probably depends on several factors.

As expected, the occurrence of a SARS-CoV-2 infection after the boost leads to a stronger immune response [9]. The maintenance or even increase of neutralizing activity against Omicron variant in patients with SARS-CoV-2 infection after the booster dose, suggests that an additional antigen exposure can enhance a memory response.

This observation is likely due to an expanded B cell repertoire and an increased affinity maturation following additional antigen exposure, as previously suggested [16,17].

The main strength of our study is the evaluation of cellular responses along with the humoral response 3 months after the booster dose as well as the positive effect of an early post-boost infection on the persistence of the immune response. The main weakness of our study is that we are not able to compare effectiveness of BNT162b2 to others mRNA vaccines or to COVID-19 vaccine combinations.

In conclusion, vaccine boost, as additional antigen exposure, provided strong response against BA.1 and BA.2 Omicron sublineages. Nevertheless, the response will likely decline over time, and is one of the factors to consider for the future of the SARS-CoV-2 vaccine strategies. The public health implication of these findings is that additional boost might be needed to maintain consistent immune response in individuals at risk of severe COVID-19 regarding the recent emergence of new sublineages (BA.4 and BA.5).

Transparency declaration

The authors declare no competing interests.

Author contributions

Conception and design of the study, data collection, analysis, writing of the manuscript, and revision of the manuscript: EKA, JD, BC-S, ML, and GL. EKA, JD, BC-S, ML, and GL take full responsibility for the integrity of the data and the accuracy of the data analysis. JL, AG, JT, DL, SM, FV, ArD, DH-G, JP, DaD, KF, DoD, LB, AlD, AS, DH, MH, FP, BA and YY participated in data analysis, and revision of the manuscript. All authors gave final approval for the version to be submitted..

Funding

This study received the Label of COVID-19 National Research Priority, awarded by the National Steering Committee on Therapeutic Trials and Other COVID-19 Research (CAPNET). This work was supported by the French government through the Programme Investissement d’Avenir (I-SITE ULNE/ANR-16-IDEX-0004 ULNE) managed by the Agence Nationale de la Recherche and was also supported by a sponsorship from GMF under the aegis of the ANRS- Maladies Infectieuses Emergentes.