key: cord-0962181-f95ext4b
authors: Zhou, Runhong; To, Kelvin Kai-Wang; Peng, Qiaoli; Chan, Jacky Man-Chun; Huang, Haode; Yang, Dawei; Lam, Bosco Hoi-Shiu; Chuang, Vivien Wai-Man; Cai, Jian-Piao; Liu, Na; Au, Ka-Kit; Tsang, Owen Tak-Yin; Yuen, Kwok-Yung; Chen, Zhiwei
title: Vaccine-breakthrough infection by the SARS-CoV-2 Omicron variant elicits broadly cross-reactive immune responses
date: 2021-12-28
journal: bioRxiv
DOI: 10.1101/2021.12.27.474218
sha: 0a196d116318c546fca45bee8883da04f8e0b5d3
doc_id: 962181
cord_uid: f95ext4b

Highly transmissible SARS-CoV-2 Omicron variant has posted a new crisis for COVID-19 pandemic control. Within a month, Omicron is dominating over Delta variant in several countries probably due to immune evasion. It remains unclear whether vaccine-induced memory responses can be recalled by Omicron infection. Here, we investigated host immune responses in the first vaccine-breakthrough case of Omicron infection in Hong Kong. We found that the breakthrough infection rapidly recruited potent cross-reactive broad neutralizing antibodies (bNAbs) against current VOCs, including Alpha, Beta, Gamma, Delta and Omicron, from unmeasurable IC50 values to mean 1:2929 at around 9-12 days, which were higher than the mean peak IC50 values of BioNTech-vaccinees. Cross-reactive spike- and nucleocapsid-specific CD4 and CD8 T cell responses were detected. Similar results were also obtained in the second vaccine-breakthrough case of Omicron infection. Our preliminary findings may have timely implications to booster vaccine optimization and preventive strategies of pandemic control.

prevention of upper airway transmission of SARS-CoV-2, the increasing numbers of 70 vaccine-breakthrough infections and re-infections have been documented 71 (Abu-Raddad et al., 2021; Birhane et al., 2021; . This situation is 72 becoming worse because of the rapid spread of SARS-CoV-2 variant of concerns 73

(VOCs) and waning of vaccine-induced immune responses (Peng et al., 2021; Wang 74 et al., 2021a; Wang et al., 2021b) . After World Health Organization (WHO) 75 designated the Omicron variant of concern (VOC) on the 26 th of November 2021, the 76 extremely rapid spread of this variant has led to another crisis of pandemic control. 77

Within a month, Omicron is replacing the Delta VOC to become the dominant 78 SARS-CoV-2 variant in many places in the South Africa, European countries and in 79

the United States (Shu and McCauley, 2017) . Two studies reported that the increased 80 risk of re-infection was associated with emergence of Omicron in South Africa and 81 Denmark (Espenhain et al., 2021; Pulliam et al., 2021) . Both vaccine-induced 82 neutralizing antibody (NAb) and current NAb combination therapy for passive 83 immunization have significantly reduced activities (Lu et al., 2021; Wang et al., 84 2021a) . Till now, it remains unclear whether vaccine-induced memory responses can 85 be recalled by the Omicron viral infection. We, therefore, investigated the host 86 immune responses in two vaccine-breakthrough cases of Omicron infection in Hong 87

Kong. Our preliminary finding of Omicron-recalled broadly cross-reactive immune 88 responses in these cases may have timely important implications to booster vaccine 89 optimization and implementing adequate preventive interventions to control the 90 pandemic. 91

On mid-November 2021, the first Chinese vaccine-breakthrough case of Omicron 93 patient (OP1) was diagnosed in a quarantine hotel in Hong Kong (Wong et al., 2021 with D614G (WT). We compared IC 50 values with 34 local vaccinees, whose blood 113 was collected around mean 30 days after the second BNT162b2-vaccination 114 (Pfizer-BioNTech) ( Figure 1A ) (Peng et al., 2021) . Consistent with recent preprint 115 publications by others, we found that the Omicron variant showed the greatest 116 resistance to BNT162b2-vaccine-induced neutralization with an average 5.9-fold 117 deficit relative to D614G ( Figure 1C) . Strikingly, however, the breakthrough 118 infection was able to elicit cross-reactive broad neutralizing antibodies (bNAbs) from 119 the unmeasurable levels (<1:20) to mean IC 50 values of 1:2929 (range 588.5-5508) 120

and from mean IC 50 1:24.3 to 1:854.5 at 9 days in OP1 and 12 days in OP2 post 121 symptoms onset (PSO), respectively ( Figure 1D) . Moreover, the amounts of NAbs 122

were consistently higher than the mean IC 50 values of BNT162b2-vaccinees across all 123

VOCs tested. In particular, there were 121.41-and 74.89-fold higher IC 50 values 124 against Beta and Omicron in OP1 than those in BNT162b2-vaccinees ( Figure 1B) . 125 Besides NAbs against the current panel of VOCs, OP1 also displayed enhanced IC 50 126 values of NAbs against 15/16 SARS-CoV-2 variants with individual mutations or 127 deletions including the E484K mutation, which conferred significant resistance to 128 vaccine-induced NAbs ( Figure S1 ). These results demonstrated that, although the 129

Omicron VOC evaded BNT162b2-vaccine-induced NAbs, the breakthrough infection 130 elicited cross-reactive bNAbs generally against all current VOCs in both OP1 and 131

OP2. 132 133 To understand cellular immune responses, we conducted flow cytometry analysis on 134

PBMCs of OP1 and OP2 collected on day 11 and 12 PSO, respectively. Multi-color 135 flow cytometry data showed no sign of severe immune suppression in both OP1 and 136 OP2 who had similar frequencies of T lymphocyte without lymphocytopenia, stable 137 conventional dendritic cell (cDC) : plasmacytoid dendritic cell (pDC) ratio and 138

normal Myeloid-derived suppressor cells (MDSCs) to mild and healthy subjects as we 139 described previously ( Figure S2 ) (Zhou et al., 2020) . For antigen-specific B cell 140 activation, we measured the frequency of Spike-specific IgG + B cells in OP1 and OP2. 141

The levels of 13.2% in OP1 and 2.31% in OP2 were relatively higher than mean 1.12% 142 (range 0.004-7.92%) found among BNT162b2-vaccinees around their peak responses 143 ( Figure 1E ). Unlike naturally infected COVID-19 patients, who display predominantly tissue-like memory (TLM) B cell response (Woodruff et al., 2020), 145 Spike-specific IgG + B cells from OP1 and OP2 exhibited the dominate phenotype of 146 resting memory (RM) (Figure 1F) , which was also found in our 147

BNT162b2-vaccinees. 148 149 Besides Spike-specific IgG + B cell responses, we measured cross-reactive T cell 150 responses to the Spike and nucleocapsid (NP) peptide pools derived from wildtype 151 SARS-CoV-2 in OP1 and OP2 as compared with BNT162b2-vaccinees by 152 intracellular cytokine staining. The cytomegalovirus (CMV) pp65 peptide pool was 153 used as a positive control. We found that Spike-and NP-specific CD4 IFN-γ 154 responses were 0.61% and 0.12% in OP1 and 0.15% and 0.10% in OP2, respectively 155 ( Figure 1G ). Moreover, Spike-and NP-specific CD8 IFN-γ responses were 0.56% 156 and 0.11% in OP1 and 0.10% and 0.08% in OP2, respectively ( Figure 1G ). These 157 results indicated that cross-reactive CD4 and CD8 T cell responses to wild type 158 SARS-CoV-2 were primarily against the Spike as compared with NP. Moreover, the 159

Spike-specific T cell responses were relatively higher in OP1 or comparable in OP2 160

as compared with mean values in BNT162b2-vaccinees (CD4 T: mean 0.19% and 161

CD8 T: mean 0.10%). Since much weaker or unmeasurable T cell responses were 162 found in severe COVID-19 patients around the same period PSO (Rydyznski 163 Moderbacher et al., 2020; Zhou et al., 2020) , T cell responses in OP1 and OP2 164

probably also contributed to disease progression control. 165

In this brief report, we provide timely communication on immune responses in two 167 cases of vaccine-breakthrough infections by the SARS-CoV-2 Omicron variant in 168

Hong Kong. Although antibody evasion has been clearly documented against 169

Omicron due to 32 amino acid changes in viral spike protein (Cameroni et al., 2021; 170 Cele et al., 2021; Planas et al., 2021a; Wang et al., 2021a) , we report here that 171

Omicron vaccine-breakthrough infections could elicit cross-reactive bNAb responses 172 against all current SARS-CoV-2 VOCs. Since the amounts of bNAb responses were 173 higher than the mean IC 50 values of in BNT162b2-vaccinees at their peak response 174 period, we believe that the Omicron infection rapidly recruited the vaccine-induced 175 memory immune responses during the acute phase of infection, which probably 176 contributed to protection and was in line with the mild clinical presentation in both 177

patients. Encouragingly, besides rapid bNAb responses, both spike-and NP-specific 178

CD4 and CD8 T cells cross-reactive to wild type peptide pools were measurable on 179 day 11-12, which probably also contributed to disease progression control (Lipsitch et 180 al., 2020; Zhou et al., 2020 

Epidemiological characterisation of the first 785 SARS-CoV-2 Omicron variant cases 331 in Denmark

T cell immune responses to SARS-CoV-2 334 and variants of concern (Alpha and Delta) in infected and vaccinated individuals

Cross-reactive memory T 337 cells and herd immunity to SARS-CoV-2

Neutralization of SARS-CoV

Omicron variant by sera from BNT162b2 or Coronavac vaccine recipients

Waning immune responses against SARS-CoV-2 among 344 vaccinees in Hong Kong

Considerable escape of 346 SARS-CoV-2 Omicron to antibody neutralization

Reduced sensitivity of 350 SARS-CoV-2 variant Delta to antibody neutralization

Safety and Efficacy of the 353 BNT162b2 mRNA Covid-19 Vaccine

Increased risk of 356 SARS-CoV-2 reinfection associated with emergence of the Omicron

Adaptive Immunity to SARS-CoV-2 in Acute COVID-19 and 361 Associations with Age and Disease Severity

GISAID: Global initiative on sharing all influenza 363 data -from vision to reality

Efficacy and safety of an 366 inactivated whole-virion SARS-CoV-2 vaccine (CoronaVac): interim results of a 367 double-blind, randomised, placebo-controlled, phase 3 trial in Turkey

Re-infection by a Phylogenetically Distinct Severe Acute Respiratory Syndrome 372

Coronavirus 2 Strain Confirmed by Whole Genome Sequencing

Safety and efficacy 376 of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim 377 analysis of four randomised controlled trials in Brazil, South Africa, and the UK

Antibody resistance of SARS-CoV-2 variants B.1.351 and 381 B.1.1.7

Transmission, viral kinetics and clinical characteristics of the 384 emergent SARS-CoV-2 Delta VOC in Guangzhou

The significant immune escape of pseudotyped SARS-CoV-2 388 variant Omicron

Variant of Concern in a designated quarantine hotel for travelers: a 391 challenge of elimination strategy of COVID-19. The Lancet Regional Health -Western 392 Pacific

Extrafollicular B cell 395 responses correlate with neutralizing antibodies and morbidity in COVID-19

Safety and immunogenicity of an inactivated SARS-CoV-2 399 vaccine, BBIBP-CorV: a randomised, double-blind, placebo-controlled, phase 1/2 trial

Acute SARS-CoV-2 Infection Impairs Dendritic Cell and 403 T Cell Responses

The authors declare no competing interests. 311