Potent cross-reactive antibodies following Omicron breakthrough in vaccinees.

Highly transmissible Omicron variants of SARS-CoV-2 currently dominate globally. Here, we compare neutralization of Omicron BA.1, BA.1.1, and BA.2. BA.2 RBD has slightly higher ACE2 affinity than BA.1 and slightly reduced neutralization by vaccine serum, possibly associated with its increased transmissibility. Neutralization differences between sub-lineages for mAbs (including therapeutics) mostly arise from variation in residues bordering the ACE2 binding site; however, more distant mutations S371F (BA.2) and R346K (BA.1.1) markedly reduce neutralization by therapeutic antibody Vir-S309. In-depth structure-and-function analyses of 27 potent RBD-binding mAbs isolated from vaccinated volunteers following breakthrough Omicron-BA.1 infection reveals that they are focused in two main clusters within the RBD, with potent right-shoulder antibodies showing increased prevalence. Selection and somatic maturation have optimized antibody potency in less-mutated epitopes and recovered potency in highly mutated epitopes. All 27 mAbs potently neutralize early pandemic strains, and many show broad reactivity with variants of concern.

SARS-CoV-2 Omicron-B.1.1.529 leads to widespread escape from neutralizing antibody responses.

On 24th November 2021, the sequence of a new SARS-CoV-2 viral isolate Omicron-B.1.1.529 was announced, containing far more mutations in Spike (S) than previously reported variants. Neutralization titers of Omicron by sera from vaccinees and convalescent subjects infected with early pandemic Alpha, Beta, Gamma, or Delta are substantially reduced, or the sera failed to neutralize. Titers against Omicron are boosted by third vaccine doses and are high in both vaccinated individuals and those infected by Delta. Mutations in Omicron knock out or substantially reduce neutralization by most of the large panel of potent monoclonal antibodies and antibodies under commercial development. Omicron S has structural changes from earlier viruses and uses mutations that confer tight binding to ACE2 to unleash evolution driven by immune escape. This leads to a large number of mutations in the ACE2 binding site and rebalances receptor affinity to that of earlier pandemic viruses.

The epitope arrangement on flavivirus particles contributes to Mab C10's extraordinary neutralization breadth across Zika and dengue viruses.

The human monoclonal antibody C10 exhibits extraordinary cross-reactivity, potently neutralizing Zika virus (ZIKV) and the four serotypes of dengue virus (DENV1-DENV4). Here we describe a comparative structure-function analysis of C10 bound to the envelope (E) protein dimers of the five viruses it neutralizes. We demonstrate that the C10 Fab has high affinity for ZIKV and DENV1 but not for DENV2, DENV3, and DENV4. We further show that the C10 interaction with the latter viruses requires an E protein conformational landscape that limits binding to only one of the three independent epitopes per virion. This limited affinity is nevertheless counterbalanced by the particle's icosahedral organization, which allows two different dimers to be reached by both Fab arms of a C10 immunoglobulin. The epitopes' geometric distribution thus confers C10 its exceptional neutralization breadth. Our results highlight the importance not only of paratope/epitope complementarity but also the topological distribution for epitope-focused vaccine design.

Evidence of escape of SARS-CoV-2 variant B.1.351 from natural and vaccine-induced sera.

The race to produce vaccines against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) began when the first sequence was published, and this forms the basis for vaccines currently deployed globally. Independent lineages of SARS-CoV-2 have recently been reported: UK, B.1.1.7; South Africa, B.1.351; and Brazil, P.1. These variants have multiple changes in the immunodominant spike protein that facilitates viral cell entry via the angiotensin-converting enzyme-2 (ACE2) receptor. Mutations in the receptor recognition site on the spike are of great concern for their potential for immune escape. Here, we describe a structure-function analysis of B.1.351 using a large cohort of convalescent and vaccinee serum samples. The receptor-binding domain mutations provide tighter ACE2 binding and widespread escape from monoclonal antibody neutralization largely driven by E484K, although K417N and N501Y act together against some important antibody classes. In a number of cases, it would appear that convalescent and some vaccine serum offers limited protection against this variant.

Antibody evasion by the P.1 strain of SARS-CoV-2.

Terminating the SARS-CoV-2 pandemic relies upon pan-global vaccination. Current vaccines elicit neutralizing antibody responses to the virus spike derived from early isolates. However, new strains have emerged with multiple mutations, including P.1 from Brazil, B.1.351 from South Africa, and B.1.1.7 from the UK (12, 10, and 9 changes in the spike, respectively). All have mutations in the ACE2 binding site, with P.1 and B.1.351 having a virtually identical triplet (E484K, K417N/T, and N501Y), which we show confer similar increased affinity for ACE2. We show that, surprisingly, P.1 is significantly less resistant to naturally acquired or vaccine-induced antibody responses than B.1.351, suggesting that changes outside the receptor-binding domain (RBD) impact neutralization. Monoclonal antibody (mAb) 222 neutralizes all three variants despite interacting with two of the ACE2-binding site mutations. We explain this through structural analysis and use the 222 light chain to largely restore neutralization potency to a major class of public antibodies.

Reduced neutralization of SARS-CoV-2 B.1.1.7 variant by convalescent and vaccine sera.

SARS-CoV-2 has caused over 2 million deaths in little over a year. Vaccines are being deployed at scale, aiming to generate responses against the virus spike. The scale of the pandemic and error-prone virus replication is leading to the appearance of mutant viruses and potentially escape from antibody responses. Variant B.1.1.7, now dominant in the UK, with increased transmission, harbors 9 amino acid changes in the spike, including N501Y in the ACE2 interacting surface. We examine the ability of B.1.1.7 to evade antibody responses elicited by natural SARS-CoV-2 infection or vaccination. We map the impact of N501Y by structure/function analysis of a large panel of well-characterized monoclonal antibodies. B.1.1.7 is harder to neutralize than parental virus, compromising neutralization by some members of a major class of public antibodies through light-chain contacts with residue 501. However, widespread escape from monoclonal antibodies or antibody responses generated by natural infection or vaccination was not observed.

The antigenic anatomy of SARS-CoV-2 receptor binding domain.

Antibodies are crucial to immune protection against SARS-CoV-2, with some in emergency use as therapeutics. Here, we identify 377 human monoclonal antibodies (mAbs) recognizing the virus spike and focus mainly on 80 that bind the receptor binding domain (RBD). We devise a competition data-driven method to map RBD binding sites. We find that although antibody binding sites are widely dispersed, neutralizing antibody binding is focused, with nearly all highly inhibitory mAbs (IC50 < 0.1 µg/mL) blocking receptor interaction, except for one that binds a unique epitope in the N-terminal domain. Many of these neutralizing mAbs use public V-genes and are close to germline. We dissect the structural basis of recognition for this large panel of antibodies through X-ray crystallography and cryoelectron microscopy of 19 Fab-antigen structures. We find novel binding modes for some potently inhibitory antibodies and demonstrate that strongly neutralizing mAbs protect, prophylactically or therapeutically, in animal models.

Reduced neutralization of SARS-CoV-2 B.1.617 by vaccine and convalescent serum.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has undergone progressive change, with variants conferring advantage rapidly becoming dominant lineages, e.g., B.1.617. With apparent increased transmissibility, variant B.1.617.2 has contributed to the current wave of infection ravaging the Indian subcontinent and has been designated a variant of concern in the United Kingdom. Here we study the ability of monoclonal antibodies and convalescent and vaccine sera to neutralize B.1.617.1 and B.1.617.2, complement this with structural analyses of Fab/receptor binding domain (RBD) complexes, and map the antigenic space of current variants. Neutralization of both viruses is reduced compared with ancestral Wuhan-related strains, but there is no evidence of widespread antibody escape as seen with B.1.351. However, B.1.351 and P.1 sera showed markedly more reduction in neutralization of B.1.617.2, suggesting that individuals infected previously by these variants may be more susceptible to reinfection by B.1.617.2. This observation provides important new insights for immunization policy with future variant vaccines in non-immune populations.

Immunogenicity of standard and extended dosing intervals of BNT162b2 mRNA vaccine.

Extension of the interval between vaccine doses for the BNT162b2 mRNA vaccine was introduced in the United Kingdom to accelerate population coverage with a single dose. At this time, trial data were lacking, and we addressed this in a study of United Kingdom healthcare workers. The first vaccine dose induced protection from infection from the circulating alpha (B.1.1.7) variant over several weeks. In a substudy of 589 individuals, we show that this single dose induces severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) neutralizing antibody (NAb) responses and a sustained B and T cell response to the spike protein. NAb levels were higher after the extended dosing interval (6-14 weeks) compared with the conventional 3- to 4-week regimen, accompanied by enrichment of CD4+ T cells expressing interleukin-2 (IL-2). Prior SARS-CoV-2 infection amplified and accelerated the response. These data on dynamic cellular and humoral responses indicate that extension of the dosing interval is an effective immunogenic protocol.

Author Correction: Human antibodies to the dengue virus E-dimer epitope have therapeutic activity against Zika virus infection.

An amendment to this paper has been published and can be accessed via a link at the top of the paper.

Broad and strong memory CD4+ and CD8+ T cells induced by SARS-CoV-2 in UK convalescent individuals following COVID-19.

The development of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) vaccines and therapeutics will depend on understanding viral immunity. We studied T cell memory in 42 patients following recovery from COVID-19 (28 with mild disease and 14 with severe disease) and 16 unexposed donors, using interferon-γ-based assays with peptides spanning SARS-CoV-2 except ORF1. The breadth and magnitude of T cell responses were significantly higher in severe as compared with mild cases. Total and spike-specific T cell responses correlated with spike-specific antibody responses. We identified 41 peptides containing CD4+ and/or CD8+ epitopes, including six immunodominant regions. Six optimized CD8+ epitopes were defined, with peptide-MHC pentamer-positive cells displaying the central and effector memory phenotype. In mild cases, higher proportions of SARS-CoV-2-specific CD8+ T cells were observed. The identification of T cell responses associated with milder disease will support an understanding of protective immunity and highlights the potential of including non-spike proteins within future COVID-19 vaccine design.

A protective Zika virus E-dimer-based subunit vaccine engineered to abrogate antibody-dependent enhancement of dengue infection.

Infections with dengue virus (DENV) and Zika virus (ZIKV) can induce cross-reactive antibody responses. Two immunodominant epitopes-one to precursor membrane protein and one to the fusion loop epitope on envelope (E) protein-are recognized by cross-reactive antibodies1-3 that are not only poorly neutralizing, but can also promote increased viral replication and disease severity via Fcγ receptor-mediated infection of myeloid cells-a process termed antibody-dependent enhancement (ADE)1,4,5. ADE is a significant concern for both ZIKV and DENV vaccines as the induction of poorly neutralizing cross-reactive antibodies may prime an individual for ADE on natural infection. In this report, we describe the design and production of covalently stabilized ZIKV E dimers, which lack precursor membrane protein and do not expose the immunodominant fusion loop epitope. Immunization of mice with ZIKV E dimers induces dimer-specific antibodies, which protect against ZIKV challenge during pregnancy. Importantly, the ZIKV E-dimer-induced response does not cross-react with DENV or induce ADE of DENV infection.

Characterization of a potent and highly unusual minimally enhancing antibody directed against dengue virus.

Dengue virus is a major pathogen, and severe infections can lead to life-threatening dengue hemorrhagic fever. Dengue virus exists as four serotypes, and dengue hemorrhagic fever is often associated with secondary heterologous infections. Antibody-dependent enhancement (ADE) may drive higher viral loads in these secondary infections and is purported to result from antibodies that recognize dengue virus but fail to fully neutralize it. Here we characterize two antibodies, 2C8 and 3H5, that bind to the envelope protein. Antibody 3H5 is highly unusual as it not only is potently neutralizing but also promotes little if any ADE, whereas antibody 2C8 has strong capacity to promote ADE. We show that 3H5 shows resilient binding in endosomal pH conditions and neutralizes at low occupancy. Immunocomplexes of 3H5 and dengue virus do not efficiently interact with Fcγ receptors, which we propose is due to the binding mode of 3H5 and constitutes the primary mechanism of how ADE is avoided.

Human antibodies to the dengue virus E-dimer epitope have therapeutic activity against Zika virus infection.

The Zika virus (ZIKV) epidemic has resulted in congenital abnormalities in fetuses and neonates. Although some cross-reactive dengue virus (DENV)-specific antibodies can enhance ZIKV infection in mice, those recognizing the DENV E-dimer epitope (EDE) can neutralize ZIKV infection in cell culture. We evaluated the therapeutic activity of human monoclonal antibodies to DENV EDE for their ability to control ZIKV infection in the brains, testes, placentas, and fetuses of mice. A single dose of the EDE1-B10 antibody given 3 d after ZIKV infection protected against lethality, reduced ZIKV levels in brains and testes, and preserved sperm counts. In pregnant mice, wild-type or engineered LALA variants of EDE1-B10, which cannot engage Fcg receptors, diminished ZIKV burden in maternal and fetal tissues, and protected against fetal demise. Because neutralizing antibodies to EDE have therapeutic potential against ZIKV, in addition to their established inhibitory effects against DENV, it may be possible to develop therapies that control disease caused by both viruses.

Germline bias dictates cross-serotype reactivity in a common dengue-virus-specific CD8+ T cell response.

Adaptive immune responses protect against infection with dengue virus (DENV), yet cross-reactivity with distinct serotypes can precipitate life-threatening clinical disease. We found that clonotypes expressing the T cell antigen receptor (TCR) ß-chain variable region 11 (TRBV11-2) were 'preferentially' activated and mobilized within immunodominant human-leukocyte-antigen-(HLA)-A*11:01-restricted CD8+ T cell populations specific for variants of the nonstructural protein epitope NS3133 that characterize the serotypes DENV1, DENV3 and DENV4. In contrast, the NS3133-DENV2-specific repertoire was largely devoid of such TCRs. Structural analysis of a representative TRBV11-2+ TCR demonstrated that cross-serotype reactivity was governed by unique interplay between the variable antigenic determinant and germline-encoded residues in the second ß-chain complementarity-determining region (CDR2ß). Extensive mutagenesis studies of three distinct TRBV11-2+ TCRs further confirmed that antigen recognition was dependent on key contacts between the serotype-defined peptide and discrete residues in the CDR2ß loop. Collectively, these data reveal an innate-like mode of epitope recognition with potential implications for the outcome of sequential exposure to heterologous DENVs.

Dengue virus sero-cross-reactivity drives antibody-dependent enhancement of infection with zika virus.

Zika virus (ZIKV) was discovered in 1947 and was thought to lead to relatively mild disease. The recent explosive outbreak of ZIKV in South America has led to widespread concern, with reports of neurological sequelae ranging from Guillain Barré syndrome to microcephaly. ZIKV infection has occurred in areas previously exposed to dengue virus (DENV), a flavivirus closely related to ZIKV. Here we investigated the serological cross-reaction between the two viruses. Plasma immune to DENV showed substantial cross-reaction to ZIKV and was able to drive antibody-dependent enhancement (ADE) of ZIKV infection. Using a panel of human monoclonal antibodies (mAbs) to DENV, we showed that most antibodies that reacted to DENV envelope protein also reacted to ZIKV. Antibodies to linear epitopes, including the immunodominant fusion-loop epitope, were able to bind ZIKV but were unable to neutralize the virus and instead promoted ADE. Our data indicate that immunity to DENV might drive greater ZIKV replication and have clear implications for disease pathogenesis and future vaccine programs for ZIKV and DENV.

A new class of highly potent, broadly neutralizing antibodies isolated from viremic patients infected with dengue virus.

Dengue is a rapidly emerging, mosquito-borne viral infection, with an estimated 400 million infections occurring annually. To gain insight into dengue immunity, we characterized 145 human monoclonal antibodies (mAbs) and identified a previously unknown epitope, the envelope dimer epitope (EDE), that bridges two envelope protein subunits that make up the 90 repeating dimers on the mature virion. The mAbs to EDE were broadly reactive across the dengue serocomplex and fully neutralized virus produced in either insect cells or primary human cells, with 50% neutralization in the low picomolar range. Our results provide a path to a subunit vaccine against dengue virus and have implications for the design and monitoring of future vaccine trials in which the induction of antibody to the EDE should be prioritized.

Oligomerization-driven avidity correlates with SARS-CoV-2 cellular binding and inhibition.

Cellular processes are controlled by the thermodynamics of the underlying biomolecular interactions. Frequently, structural investigations use one monomeric binding partner, while ensemble measurements of binding affinities generally yield one affinity representative of a 1:1 interaction, despite the majority of the proteome consisting of oligomeric proteins. For example, viral entry and inhibition in SARS-CoV-2 involve a trimeric spike surface protein, a dimeric angiotensin-converting enzyme 2 (ACE2) cell-surface receptor and dimeric antibodies. Here, we reveal that cooperativity correlates with infectivity and inhibition as opposed to 1:1 binding strength. We show that ACE2 oligomerizes spike more strongly for more infectious variants, while exhibiting weaker 1:1 affinity. Furthermore, we find that antibodies use induced oligomerization both as a primary inhibition mechanism and to enhance the effects of receptor-site blocking. Our results suggest that naive affinity measurements are poor predictors of potency, and introduce an antibody-based inhibition mechanism for oligomeric targets. More generally, they point toward a much broader role of induced oligomerization in controlling biomolecular interactions.

Heterologous versus homologous COVID-19 booster vaccination in previous recipients of two doses of CoronaVac COVID-19 vaccine in Brazil (RHH-001): a phase 4, non-inferiority, single blind, randomised study.

INTRODUCTION: The inactivated whole-virion SARS-CoV-2 vaccine (CoronaVac, Sinovac) has been widely used in a two-dose schedule. We assessed whether a third dose of the homologous or a different vaccine could boost immune responses. METHODS: RHH-001 is a phase 4, participant masked, two centre, safety and immunogenicity study of Brazilian adults (18 years and older) in São Paulo or Salvador who had received two doses of CoronaVac 6 months previously. The third heterologous dose was of either a recombinant adenoviral vectored vaccine (Ad26.COV2-S, Janssen), an mRNA vaccine (BNT162b2, Pfizer-BioNTech), or a recombinant adenoviral-vectored ChAdOx1 nCoV-19 vaccine (AZD1222, AstraZeneca), compared with a third homologous dose of CoronaVac. Participants were randomly assigned (5:6:5:5) by a RedCAP computer randomisation system stratified by site, age group (18-60 years or 61 years and over), and day of randomisation, with a block size of 42. The primary outcome was non-inferiority of anti-spike IgG antibodies 28 days after the booster dose in the heterologous boost groups compared with homologous regimen, using a non-inferiority margin for the geometric mean ratio (heterologous vs homologous) of 0·67. Secondary outcomes included neutralising antibody titres at day 28, local and systemic reactogenicity profiles, adverse events, and serious adverse events. This study was registered with Registro Brasileiro de Ensaios Clínicos, number RBR-9nn3scw. FINDINGS: Between Aug 16, and Sept 1, 2021, 1240 participants were randomly assigned to one of the four groups, of whom 1239 were vaccinated and 1205 were eligible for inclusion in the primary analysis. Antibody concentrations were low before administration of a booster dose with detectable neutralising antibodies of 20·4% (95% CI 12·8-30·1) in adults aged 18-60 years and 8·9% (4·2-16·2) in adults 61 years or older. From baseline to day 28 after the booster vaccine, all groups had a substantial rise in IgG antibody concentrations: the geometric fold-rise was 77 (95% CI 67-88) for Ad26.COV2-S, 152 (134-173) for BNT162b2, 90 (77-104) for ChAdOx1 nCoV-19, and 12 (11-14) for CoronaVac. All heterologous regimens had anti-spike IgG responses at day 28 that were superior to homologous booster responses: geometric mean ratios (heterologous vs homologous) were 6·7 (95% CI 5·8-7·7) for Ad26.COV2-S, 13·4 (11·6-15·3) for BNT162b2, and 7·0 (6·1-8·1) for ChAdOx1 nCoV-19. All heterologous boost regimens induced high concentrations of pseudovirus neutralising antibodies. At day 28, all groups except for the homologous boost in the older adults reached 100% seropositivity: geometric mean ratios (heterologous vs homologous) were 8·7 (95% CI 5·9-12·9) for Ad26.COV2-S vaccine, 21·5 (14·5-31·9) for BNT162b2, and 10·6 (7·2-15·6) for ChAdOx1 nCoV-19. Live virus neutralising antibodies were also boosted against delta (B.1.617.2) and omicron variants (B.1.1.529). There were five serious adverse events. Three of which were considered possibly related to the vaccine received: one in the BNT162b2 group and two in the Ad26.COV2-S group. All participants recovered and were discharged home. INTERPRETATION: Antibody concentrations were low at 6 months after previous immunisation with two doses of CoronaVac. However, all four vaccines administered as a third dose induced a significant increase in binding and neutralising antibodies, which could improve protection against infection. Heterologous boosting resulted in more robust immune responses than homologous boosting and might enhance protection. FUNDING: Ministry of Health, Brazil.

Reactogenicity and immunogenicity after a late second dose or a third dose of ChAdOx1 nCoV-19 in the UK: a substudy of two randomised controlled trials (COV001 and COV002).

BACKGROUND: COVID-19 vaccine supply shortages are causing concerns about compromised immunity in some countries as the interval between the first and second dose becomes longer. Conversely, countries with no supply constraints are considering administering a third dose. We assessed the persistence of immunogenicity after a single dose of ChAdOx1 nCoV-19 (AZD1222), immunity after an extended interval (44-45 weeks) between the first and second dose, and response to a third dose as a booster given 28-38 weeks after the second dose. METHODS: In this substudy, volunteers aged 18-55 years who were enrolled in the phase 1/2 (COV001) controlled trial in the UK and had received either a single dose or two doses of 5 × 1010 viral particles were invited back for vaccination. Here we report the reactogenicity and immunogenicity of a delayed second dose (44-45 weeks after first dose) or a third dose of the vaccine (28-38 weeks after second dose). Data from volunteers aged 18-55 years who were enrolled in either the phase 1/2 (COV001) or phase 2/3 (COV002), single-blinded, randomised controlled trials of ChAdOx1 nCoV-19 and who had previously received a single dose or two doses of 5 × 1010 viral particles are used for comparison purposes. COV001 is registered with ClinicalTrials.gov, NCT04324606, and ISRCTN, 15281137, and COV002 is registered with ClinicalTrials.gov, NCT04400838, and ISRCTN, 15281137, and both are continuing but not recruiting. FINDINGS: Between March 11 and 21, 2021, 90 participants were enrolled in the third-dose boost substudy, of whom 80 (89%) were assessable for reactogenicity, 75 (83%) were assessable for evaluation of antibodies, and 15 (17%) were assessable for T-cells responses. The two-dose cohort comprised 321 participants who had reactogenicity data (with prime-boost interval of 8-12 weeks: 267 [83%] of 321; 15-25 weeks: 24 [7%]; or 44-45 weeks: 30 [9%]) and 261 who had immunogenicity data (interval of 8-12 weeks: 115 [44%] of 261; 15-25 weeks: 116 [44%]; and 44-45 weeks: 30 [11%]). 480 participants from the single-dose cohort were assessable for immunogenicity up to 44-45 weeks after vaccination. Antibody titres after a single dose measured approximately 320 days after vaccination remained higher than the titres measured at baseline (geometric mean titre of 66·00 ELISA units [EUs; 95% CI 47·83-91·08] vs 1·75 EUs [1·60-1·93]). 32 participants received a late second dose of vaccine 44-45 weeks after the first dose, of whom 30 were included in immunogenicity and reactogenicity analyses. Antibody titres were higher 28 days after vaccination in those with a longer interval between first and second dose than for those with a short interval (median total IgG titre: 923 EUs [IQR 525-1764] with an 8-12 week interval; 1860 EUs [917-4934] with a 15-25 week interval; and 3738 EUs [1824-6625] with a 44-45 week interval). Among participants who received a third dose of vaccine, antibody titres (measured in 73 [81%] participants for whom samples were available) were significantly higher 28 days after a third dose (median total IgG titre: 3746 EUs [IQR 2047-6420]) than 28 days after a second dose (median 1792 EUs [IQR 899-4634]; Wilcoxon signed rank test p=0·0043). T-cell responses were also boosted after a third dose (median response increased from 200 spot forming units [SFUs] per million peripheral blood mononuclear cells [PBMCs; IQR 127-389] immediately before the third dose to 399 SFUs per milion PBMCs [314-662] by day 28 after the third dose; Wilcoxon signed rank test p=0·012). Reactogenicity after a late second dose or a third dose was lower than reactogenicity after a first dose. INTERPRETATION: An extended interval before the second dose of ChAdOx1 nCoV-19 leads to increased antibody titres. A third dose of ChAdOx1 nCoV-19 induces antibodies to a level that correlates with high efficacy after second dose and boosts T-cell responses. FUNDING: UK Research and Innovation, Engineering and Physical Sciences Research Council, National Institute for Health Research, Coalition for Epidemic Preparedness Innovations, National Institute for Health Research Oxford Biomedical Research Centre, Chinese Academy of Medical Sciences Innovation Fund for Medical Science, Thames Valley and South Midlands NIHR Clinical Research Network, AstraZeneca, and Wellcome.