Bivalent and Monovalent SARS-CoV-2 Variant Vaccine Boosters Improve coverage of the known Antigenic Landscape: Results of the COVID-19 Variant Immunologic Landscape (COVAIL) Trial

Vaccine protection against COVID-19 wanes over time and has been impacted by the emergence of new variants with increasing escape of neutralization. The COVID-19 Variant Immunologic Landscape (COVAIL) randomized clinical trial (clinicaltrials.gov NCT 05289037) compares the breadth, magnitude and durability of antibody responses induced by a second COVID-19 vaccine boost with mRNA (Moderna mRNA-1273 and Pfizer-BioNTech BNT 162b2), or adjuvanted recombinant protein (Sanofi CoV2 preS DTM-AS03) monovalent or bivalent vaccine candidates targeting ancestral and variant SARS-CoV-2 spike antigens (Beta, Delta and Omicron BA.1). We found that boosting with a variant strain is not associated with loss in neutralization against the ancestral strain. However, while variant vaccines compared to the prototype/wildtype vaccines demonstrated higher neutralizing activity against Omicron BA.1 and BA.4/5 subvariants for up to 3 months after vaccination, neutralizing activity was lower for more recent Omicron subvariants. Our study, incorporating both antigenic distances and serologic landscapes, can provide a framework for objectively guiding decisions for future vaccine updates.

In this adaptive phase 2 clinical trial, we evaluated boosting with ancestral and variant SARS-CoV-2 spike protein(s) (Beta, Delta and Omicron BA.1), alone or in combination, using both mRNA vaccines (Moderna and P zer BioNTech mRNA), and recombinant protein vaccine (Sano AS03-adjuvanted), to assess the breadth, magnitude and durability of neutralizing antibody responses.

Study Population
From March 30 to May 6, 2022, 602 participants were randomized and 597 received a Moderna mRNA vaccine in stage 1 ( Table 1, Supplemental Figure   S15). 21 From May 12 to 27, 2022, 313 participants were randomized and 312 received a P zer BioNTech mRNA vaccine in stage 2 (Table 1 and Supplemental Figure S16). From June 8 to 17,2022, 153 participants were randomized and 152 received a Sano protein vaccine in stage 3 (Table 1 and Supplemental Figure S17). Baseline demographics were similar across between study arms within each stage (Table 1). Median age was 53 years (range: 18-85) for stage 1, 47 years (range: 20-83) for stage 2 and 45 years (range: 18-79) for stage 3; 35%, 30% and 20% were ≥ 65 years for each stage, respectively. The majority of participants were female (53-59%); 6 to 11% were Hispanics and 73 to 82% were White. Most participants (94-100% per arm) had received an mRNAbased primary series and initial boost vaccination. Twenty percent, 33% and 41% in stages 1, 2 and 3, respectively, were de ned as previously infected based on anti-N antibody seropositivity at baseline and/or by self-reported past positive SARS-CoV-2 PCR or antigen testing. Median duration (range) between study vaccination and the last previous vaccination or infection was 168 (110-333) days, 198 (106-333) days and 197 (79-359) days for stages 1, 2 and 3, respectively. Median follow up duration at data cutoff was 228 days, 193 days and 176 days for stages 1, 2 and 3 respectively.

Safety
The frequency and severity of solicited local and systemic adverse events (AEs) after vaccination were similar to other booster trials 22 and did not differ between arms in each stage (Supplemental Figs. 1-6). The most frequently reported solicited local AE was injection-site pain (83% of participants for stage 1, 77% for stage 2, 74% for stage 3). The most common solicited systemic AEs were fatigue (50-67%) and myalgia (39-57%). Most solicited AEs were mild to moderate with only 0-1% severe local AEs and 0.7-4% severe systemic AEs. A summary of all AEs is presented in Supplemental Figs. 7-9. As of the data cutoff, 13 participants in stage 1, 4 participants in stage 2, and 1 participant in stage 3 had a serious AE (SAE); all were deemed unrelated to study product. There was one related AESI in stage 1 of a young man who reported chest pain 1 day after vaccination that was initially evaluated as possible myocarditis, which was ultimately excluded due to a normal troponin I level and normal cardiac MRI. There was one death unrelated to study product due to cardiac arrest from advanced coronary artery disease.  Fig. 10).
For uninfected participants, all Omicron BA.1 containing vaccines (Day 29 GMT D614G between 11,963 and 16,001) boosted PsVN Ab to D614G similarly to the Prototype vaccine (Day 29 GMT D614G = 12,600). (Fig. 1 suggesting either differences in antibody maturation or antigenic distance among the variants versus a ceiling with D614G. The antibody responses with Omicron BA.1-containing vaccines were more durable, with a smaller geometric mean fold decline (GMFD) from Day 29 to Day 91 for Omicron subvariants (GMFD BA.1 =2.0 to 2.   Table 13). The assays were performed in a separate laboratory using a pseudovirus platform that resembles but is not identical to the one used for the other datasets in this study.
PsVN Ab were highest against the ancestral D614G variant in both groups. Higher GMT estimates against all Omicron subvariants were observed at Day 15 with the Omicron BA.1 + Prototype bivalent vaccine when compared to the Prototype. More pronounced immune escape was seen with the recent variants (BQ.1.1 and XBB.1). The PsVN Ab response was more durable with the bivalent compared to the Prototype vaccine at Day 91 relative to Day 15 with a GMFD  (Table 1). Neutralizing antibodies were assessed with the same assay used for the main dataset in stage 1.
Consistent with stage 1 results involving a similar mRNA vaccine technology, PsVN Ab peaked on Day 15, remained relatively stable on Day 29, were similar between older (≥ 65 years) and younger adults, and were 2-4 times higher in previously infected participants compared to uninfected participants (Supplemental Tables 9 and 10 For uninfected participants (Supplemental Table 9  Neutralizing Antibody Responses for Stage 3 (Sano AS03-adjuvanted protein) Stage 3 participants were boosted with one of three Sano adjuvanted recombinant spike protein vaccine products, including the prototype vaccine, a monovalent Beta vaccine, and a bivalent Beta + Prototype vaccine (Table 1). No product containing Omicron spike protein was available at the time the study was conducted and Day 15 samples were not tested for stage 3 arms. Neutralizing antibodies were assessed on Days 1, 29 and 91 in the same assay used for the main datasets in stages 1 and 2.
PsVN Ab at Day 29 after vaccination with Sano variant vaccines were similar between older (≥ 65 years) and younger adults. Day 29 PsVN Ab titers were approximately 4-5 times higher in previously infected compared to uninfected recipients of the monovalent Beta and monovalent Protytpe vaccines arms but there was less numerical difference in Day 29 titers (1.5-2 times higher) between previously infected and uninfected recipients of the bivalent Beta + Prototype vaccine. (Fig. 1, Supplemental Tables 11 and 12 Table 11 and Supplemental Fig. 12 Table 15). The Day 91 GMRs in stages 1 and 2 were similar or higher to those observed for D29. In stage 3, all Beta-containing Sano vaccines led to a Day 91 GMR of greater than 1 relative to the Prototype vaccine, though the unadjusted lower bound con dence interval did cross 1 (Supplemental Table 16).

Antigenic Cartography and Antibody Landscapes
The antigenic landscapes for each vaccine arm across the 3 stages (Fig. 3A) were derived based on the antigenic map base map by Wilks et. al. 23 . Figure 3B shows the GMT antibody landscapes for each vaccine arm in the 3 stages strati ed by prior infection, with the corresponding neutralizing antibody titers above the variant's map position.
All vaccine arms for each of their respective stages in uninfected participants had similar pre-vaccination antigenic landscapes, with the apex over D614G, as expected (Fig. 3B). After vaccination, all arms, in all 3 stages, had antibody titers that raised and attened the antigenic landscape. In uninfected cohorts for all 3 stages, variant-containing vaccines lifted titers against BA.1 and BA.4/5 and attened the landscapes better than the Prototype or Wildtype vaccines. A second booster dose raised antibody titers in uninfected participants to the titers observed in previously infected participants at baseline (Fig. 3B) (Supplemental Figs. 13 and 14).

SARS-CoV-2 infections
There were 267 self-reported COVID-19 illnesses, occurring after randomization among 973 participants in single dose arms by data cutoff, 1 of which resulted in a brief hospitalization, lasting less than 24 hours, due to hypoxemia. The incidence of infections in this trial re ect the community transmission, with the majority occurring during the Omicron BA.5 wave in the United States. At any point in time, participants from different stages will be in different points in follow up. Therefore assessing incidence across stages is not a valid comparison. Kaplan

DISCUSSION
The continued emergence of SARS-CoV-2 VOCs led to a recommendation to update COVID-19 vaccines. 24 The strains selected in 2022 for modi ed vaccines covered circulating strains at the time of vaccine development, not necessarily variants that would drift antigenically from Omicron BA.1 and BA.4/5 or evolve from other distinct locations on the phylogenetic tree. Therefore, it is important to investigate not only immune responses to known variants, but also the antigenic relationships among different SARS-CoV-2 VOCs 25 and how variant vaccines may alter immunologic landscapes to cover antigenic areas where new strains may emerge. Here, we described the magnitude, breadth, and landscapes of the neutralizing antibody response following a second booster with investigational monovalent and bivalent variant-speci c vaccines re ective of the diverse SARS-CoV-2 immunologic background seen in the general population and utilizing different vaccine platforms.
Our ndings support that mRNA and adjuvanted protein variant vaccines elicit substantial cross-reactive neutralizing antibodies to D614G as well as to Though speci c correlates of protection for infection with recent Omicron subvariants are not well understood, neutralizing antibody titers have been used to infer protection during the D614G wave of the pandemic, when the circulating virus closely matched the vaccine strain, 32 and the resulting immunologic data has served as the basis for emergency use authorization for booster vaccines by regulatory agencies. 33,34 The improved serologic response with variant containing vaccines over Prototype/Wildtype vaccines in our study and others [35][36][37] provide evidence that broad cross-protection may be conferred without a variant-chasing approach and warrants further mechanistic exploration.
For all vaccine candidates, including vaccine products not containing Prototype, the antibody titers were higher against D614G compared to the VOCs, supporting the hypothesis of back-boosting to the ancestral strain seen in previous studies. [36][37][38] This suggests that future generations of SARS-CoV-2 vaccines may be able to omit Prototype or Wildtype sequences without losing the ability to neutralize D614G, or other variants within close antigenic distance, in people who previously received the Prototype vaccines. Furthermore, Omicron BA.1 or Beta monovalent vaccines were nearly equivalent to Omicron BA.1 + Prototype or Beta + Prototype bivalent vaccines for neutralization of B.1.351 and both Omicron subvariants (BA.1 and BA.4/5) further supporting the premise that monovalent variant vaccines could replace bivalent vaccines as the updated boost in the future. 30 Notably, although variant vaccines improved neutralizing activity against Omicron subvariants, these titers decreased for more recent Omicron subvariants.
While the ID 50 against BA.1 and BA.4/5 remained high, the neutralization titers for subvariants BQ.1.1 and XBB.1 were much lower. Additionally, we noted a high rate of infections which occurred during the BA.4/5 wave and subsequent waves with XBB.1 and BQ.1.1. These infections occurred more frequently in previously uninfected compared to previously infected individuals highlighting the importance of hybrid immunity in protection against disease. 31 In addition, infections occurred in younger rather than older adults likely re ecting behavioral differences impacting risk of exposure. Our study was not designed to assess vaccine effectiveness (VE). Though recent data suggest possibly higher VE against Omicron subvariants with bivalent vaccine boosts (Prototype + Omicron BA.4/5, Prototype + Omicron BA.1) compared with the Prototype vaccine, 18,39 our ndings highlight concerns that variant vaccines are unlikely to keep pace with virus evolution and that other immune correlates of protection beyond antibody responses need to be explored.
Our study has several limitations. First, the sample size is small for certain subgroups of interest such prior infection (27%) and adults older than 65 years (31%). Second, T cell responses and antibody effector functions, which may be critical to preventing severe disease, 40 have not yet been evaluated.
Additionally, clonal and kinetic analyses of the memory B cell response, while underway, are not available to further differentiate the durability of the antibody response elicited by variant-containing vaccines. Finally, participants were only randomized to different arms within each stage and not between stages which enrolled sequentially at different calendar times leading to different exposures to circulating variants prior to and after enrollment. This precludes head-tohead comparisons of rates of infections or neutralization titers across stages.
In conclusion, these data demonstrate that updating vaccines to target recent variants provides modestly improved and broadly cross-protective neutralizing antibody responses against diverse SARS-CoV-2 variants without sacri cing boosting immunity to the ancestral strain. The precise degree to which the enhanced antibody response elicited by updated vaccines will restore protection against disease after infection with heterologous or homologous strains needs further con rmation by real-world effectiveness studies. Our study incorporating both antigenic distances and serologic landscapes serve as a framework for objectively guiding decisions for future vaccine updates.

Study Design and Eligibility Criteria
This phase 2 open-label, randomized, clinical trial was performed at 22 sites in the US (Supplemental Table 1). Eligible participants were healthy adults 18 years of age and older (with or without a history of prior SARS-CoV-2 infection) who had received a primary series and a single homologous or heterologous boost with an approved or emergency use authorized COVID-19 vaccine (Supplemental Table 2). The most recent vaccine dose, and/or prior infection must have occurred at least 16 weeks prior to randomization. Full eligibility criteria are described at clinicaltrials.gov (NCT 05289037).
Eligible participants were strati ed by age (18-64 and ≥ 65 years) and history of con rmed SARS-CoV-2 infection, and randomly assigned across arms within each stage in an equal ratio using block randomization methodology with blocks of size 6 and 12 for stages 1 and 2 and blocks of size 3 and 6 for stage 3. Subjects were randomized using the Advantage eClinical system used by the Statistical Data Coordinating Center. As this was an unblinded study, no effort was made to conceal the assignment post randomization. Sample size was chosen to be able to detect common adverse events and estimate immunogenicity parameters with acceptable precision (See protocol for further details). After providing informed consent, participants underwent screening, including con rmation of COVID-19 vaccination history, medical history, a targeted physical examination, and a urine pregnancy test (if indicated). Safety and immunogenicity assessments were performed on Days 1, 15 and 29, and at 3, 6, 9 and 12 months after last vaccination. Although the study was not designed to evaluate booster vaccine effectiveness, we collected information on antigen or PCR-con rmed symptomatic or asymptomatic SARS-CoV-2 infection at any time after randomization. A nasal swab sample was collected for viral sequencing in persons testing positive. Immunologic data are currently available up to the Day 91 visit after rst vaccination. The safety data cutoff was December 2, 2022.
The trial was reviewed and approved by a central institutional review board and overseen by an independent Data and Safety Monitoring Board. The trial was sponsored and funded by the National Institutes of Health (NIH). The NIAID SARS-CoV-2 Assessment of Viral Evolution (SAVE) program team was consulted to inform study arm design and variant vaccine selection.

Trial vaccines
Trial vaccines are listed in Table 1 and Supplemental Table 3. Trial vaccines were provided by Moderna (Cambridge, MA) for stage 1 (50 mcg per vaccine), P zer BioNTech (New York, NY) for stage 2 ( 30 mcg per vaccine) and Sano (Paris, France) for stage 3 (5 mcg per vaccine). The vaccine candidates were manufactured similarly to their corresponding authorized or approved vaccines in the US or Europe.

Study outcomes
The primary objective was to evaluate humoral immune responses of candidate SARS-CoV-2 variant vaccines, alone or in combination. The secondary objective was to evaluate the safety of candidate SARS-CoV-2 variant vaccines assessed by solicited injection site and systemic adverse events (AEs), which were collected for 7 days after vaccination; unsolicited AEs through Day 29; and serious adverse events (SAEs), new-onset chronic medical conditions (NOCMCs), adverse events of special interest (AESIs), AEs leading to withdrawal, and medically attended adverse events (MAAEs) through the duration of the trial.
Exploratory objectives included sequencing strains from infections for variant spike lineage and assessing anti-nucleocapsid serology. Information on antigenor PCR-con rmed symptomatic or asymptomatic SARS-CoV-2 infection at any time after randomization was collected.

Statistical analysis
The primary objective of this study is to evaluate the magnitude, breadth and durability of SARS-CoV-2 speci c antibody titers in serum samples by estimating 95% con dence intervals (CI) for the geometric mean titer (GMT) at each timepoint when samples are collected. No pre-speci ed formal hypothesis tests were planned. The geometric mean fold rise (GMFR) is calculated as the geometric mean of titers at a timepoint divided by titers at Day 1. The geometric mean ratio to D614G (GMR D614G ) is the geometric mean of the ratio of D614G titers against titers for a variant of concern. Seropositive rate is calculated as the proportion of participants with titers above the lower limit of detection (LLOD). 95% CI for GMT, GMFR, and GMR D614G are calculated using the Student's tdistribution and 95% CI for seropositive rate is calculated using the Clopper-Pearson binomial method. For the purpose of analysis, participants were de ned as previously infected by self-report of a con rmed positive antigen or PCR testing or the detection of anti-SARS-CoV-2 N antibodies. Participants with a SARS-CoV-2 infection occurring between vaccination and a pre-speci ed immunogenicity timepoint were excluded from immunogenicity analysis at that timepoint and thereafter.
ANCOVA models were used to estimate GMT ratios of variant vaccines compared to Prototype vaccine and included independent variables for vaccination arms, age (18-64 years and ≥ 65 years of age), previous infection history, and baseline titers. For modeling purposes, titers were log 10 transformed and estimated mean differences were back transformed to generate GMT ratios between vaccination groups. Unadjusted 97.5% con dence intervals based on the t-distribution are reported.
Infection rates are estimated using Kaplan-Meier methodology.
All analyses were done in SAS v9.4 or R v4.2.2 or higher.

Antigenic cartography and antibody landscapes
Antigenic cartography uses antibody neutralization data to position virus variants and sera relative to each other in an n-dimensional Euclidean space, in this case a 2-dimensional space, as previously described. 43,44 The distance between variants can be understood as a measure of antigenic similarity. Brie y, for each serum-variant pair, the fold-change from the maximum titer variant in the speci c serum is calculated to obtain a target distance from the serum. Serum and variant coordinates are then optimized such that difference between Euclidean map distance and this target distance is minimized, with one map unit corresponding to one two-fold dilution of neutralization titers on the log2 scale. Here, the antigenic map published in Wilks et al. 44 was used as basis for the antibody landscapes, where neutralization titers against virus variants are plotted in a third dimension above the corresponding variant in an antigenic map and a continuous surface is tted to these titers. 23 Antibody landscapes were constructed using the ablandscape. t function 45,46 of the ablandscape package

Declarations Data availability
All data is included in the manuscript. The protocol for the study is provided as supplementary materials.

Code availability
The code for antibody landscapes and titer line plots is publicly available in a GitHub repository (https://github.com/acorg/branche_et_al2023). Table   Table 1 is available in the Supplementary Files section.