Bivalent Coronavirus Disease 2019 Vaccine Antibody Responses to Omicron Variants Suggest That Responses to Divergent Variants Would Be Improved With Matched Vaccine Antigens

Abstract

We compared neutralizing antibody responses to BA.4/5, BQ.1.1, XBB, and XBB.1.5 Omicron severe acute respiratory syndrome coronavirus 2 variants after a bivalent or ancestral coronavirus disease 2019 (COVID-19) messenger RNA booster vaccine or postvaccination infection. We found that the bivalent booster elicited moderately high antibody titers against BA.4/5 that were approximately 2-fold higher against all Omicron variants than titers elicited by the monovalent booster. The bivalent booster elicited low but similar titers against both XBB and XBB.1.5 variants. These findings inform risk assessments for future COVID-19 vaccine recommendations and suggest that updated COVID-19 vaccines containing matched vaccine antigens to circulating divergent variants may be needed.

Bivalent mRNA COVID-19 booster vaccines elicit 2-fold higher antibody titers against Omicron variants than monovalent ancestral vaccines, but titers against XBB and XBB.1.5 were low. The findings suggest that responses to divergent variants would be improved with matched vaccine antigens.

The evolution of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and dynamic population immunity from combinations of vaccines and infections present challenges for vaccination strategies. Since the Omicron BA.5 subvariant became dominant in mid-2022, additional immune-evasive Omicron variants emerged, including BA.5 lineage BQ.1 and BQ.1.1 variants and BA.2 lineage recombinant variants XBB, XBB.1, and XBB.1.5 (Figure 1A). Bivalent messenger RNA (mRNA) coronavirus disease 2019 (COVID-19) vaccines encoding both ancestral and Omicron BA.4/5 spike proteins have been used to boost immunity since September 2022. However, there are concerns that prior monovalent vaccinations encoding the ancestral variant may hinder antibody responses to new variants through immune imprinting [1, 2]. Recommendations about the timing of COVID-19 vaccine boosters and variant composition of updated COVID-19 vaccines are based on risk assessments that consider preexisting immunity. Understanding how bivalent boosting compares to monovalent (ancestral) boosting or a post-vaccination infection (PVI) in eliciting neutralizing antibodies to recent circulating Omicron variants, including XBB and XBB.1.5, informs risk assessments for decisions about use of vaccine boosters or vaccine composition updates [3].

Figure 1.

Neutralization of Omicron subvariants by postvaccination and postvaccination infection serum samples. A, Amino acid mutations and deletions (Del) in spike proteins of ancestral (D614G listed here) and recently emerged Omicron subvariants are indicated in reference to USA-WA1/2020. Shaded boxes indicate an amino acid substitution relative to WA1/2020. Amino acid substitutions, indicated by their single-letter abbreviation, are listed in the shaded box for variants that have different substitutions in those positions. N-terminal domain and receptor-binding domain in S1 are marked. BA.4 and BA.5 have the identical spike sequence; thus, their spikes are marked as BA.4/5. B, Neutralizing antibody titers against the indicated variants in human serum samples after different exposures by messenger RNA coronavirus disease 2019 vaccines and postvaccination infections were measured in lentiviral-based pseudovirus neutralization assays. Geometric mean titers (GMTs) assessed as the reciprocal dilution of serum that neutralizes 50% of the input pseudovirus (ID₅₀) against different variant pseudoviruses were compared. Serum samples in which the ID₅₀ fell below the limit of detection at 1:40 dilution were assigned an ID₅₀ value of 20. Dots indicate results from individual participants, and bars indicate GMT with 95% confidence interval. P < .05 was considered statistically significant. All neutralization titers were log₂ transformed for analyses. *P < .05. C, Neutralizing antibody titers (GMT) in the V4 (4 doses of the monovalent (ancestral) vaccine), V3 + Bi (3 doses of the monovalent (ancestral) vaccine and 1 dose of the bivalent (ancestral and BA.4/5) vaccine), V3 + PVI (3 doses of the monovalent (ancestral) vaccine and 1 post-vaccination infection), and V4 + Bi (4 doses of the monovalent (ancestral) vaccine and 1 dose of the bivalent (ancestral and BA.4/5) vaccine) groups are illustrated with titers against D614G, BA.4/5, BQ.1.1, XBB, and XBB.1.5. D, Neutralizing antibody titers (GMT) against D614G, BA.4/5, BQ.1.1, XBB, and XBB.1.5 are compared in the V4, V3 + Bi, V3 + PVI, and V4 + Bi groups.

METHODS

Study Cohort

Sera were collected from persons with different exposure histories in a well-characterized, prospective cohort in the Prospective Assessment of SARS-CoV-2 Seroconversion (PASS) study (Supplementary Figure 1 and Supplementary Table 1). Details of the PASS study protocol, including details of the inclusion and exclusion criteria, have been published [4]. The PASS study (Protocol IDCRP-126) was approved by the Uniformed Services University of the Health Sciences Institutional Review Board in compliance with all applicable federal regulations governing the protection of human participants. All study participants provided informed consent. Details of vaccination information are summarized in Supplementary Tables 2 and 3. Sera from individuals who received 4 monovalent (ancestral) mRNA COVID-19 vaccines (V4, n = 16) were collected a median of 33 days (interquartile range [IQR], 22.5–53.0 days) after last vaccination during 2 May 2022 to 3 November 2022. Sera from individuals who received 3 monovalent (ancestral) mRNA COVID-19 vaccines and 1 bivalent (ancestral and BA.4/5) mRNA COVID-19 vaccine (V3 + Bi, n = 19) were collected a median of 29 days (IQR, 27.5–35.0 days) after last vaccination during 18 October 2022 to 16 November 2022. Sera from individuals who received 3 monovalent (ancestral) mRNA COVID-19 vaccines before a PVI during the BA.1 wave (V3 + PVI, n = 25) were collected a median of 56 days (IQR, 37.0–75.0 days) after infection during 2 February 2022 to 19 November 2022. We also collected serum from persons who received 4 doses of monovalent (ancestral) mRNA COVID-19 vaccine before 1 bivalent (ancestral and BA.4/5) mRNA COVID-19 vaccine (V4 + Bi, n = 8) a median of 37 days (IQR, 32.8–43.0 days) after last vaccination during 20 October 2022 to 16 November 2022.

Pseudovirus Production and Neutralization Assay

Human immunodeficiency virus (HIV)–based lentiviral pseudoviruses with desired SARS-CoV-2 spike proteins (D614G [EPI_ISL_5851484], BA.4/5 [EPI_ISL_12464782], BQ.1.1 [EPI_ISL_16364753], XBB [EPI_ISL_16160901] obtained from National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, and XBB.1.5 [EPI_ISL_15687648] kindly provided by Dr Shaunna Shen from Duke University, Durham, North Carolina) were generated as previously described [5]. Pseudoviruses were produced in 293 T cells by cotransfection of 5 μg of pCMVΔR8.2, 5 μg of pHR’CMVLuc, and 0.5 μg of pVRC8400 encoding a codon-optimized spike gene. Pseudovirus supernatants were collected approximately 48 hours posttransfection, filtered through a 0.45-μm low protein binding filter, and stored at −80°C. Pseudovirus neutralization assays were performed using 293T-ACE2-TMPRSS2 cells in 96-well plates [5]. Pseudoviruses with titers of approximately 10⁶ relative luminescence units per milliliter (RLU/mL) of luciferase activity were incubated with serially diluted sera for 2 hours at 37°C prior to inoculation onto the plates that were preseeded 1 day earlier with 3.0 × 10⁴ cells per well. Pseudovirus infectivity was determined 48 hours postinoculation for luciferase activity by luciferase assay reagent (Promega) according to the manufacturer's instructions. The inverse of the sera dilutions causing a 50% reduction of RLU compared to control was reported as the neutralization titer (ID₅₀). Titers were calculated using a nonlinear regression curve fit (GraphPad Prism Software, La Jolla, California). The mean titer from at least 2 independent experiments each with intra-assay duplicates was reported as the final titer.

Statistical Analysis

Mann–Whitney test was used for 2-group comparisons and Tukey multiple comparisons test for multiple groups, and geometric mean titers (GMTs) with 95% confidence intervals (CIs) were determined using GraphPad Prism software. P values < .05 were considered statistically significant. All neutralization titers were log₂ transformed for analyses.

RESULTS

The primary outcome assessed in this study was neutralizing titers between V4 and V3 + Bi groups. Neutralizing antibody titers in individuals who received a fourth vaccine dose as bivalent vaccine (V3 + Bi) were approximately 2-fold greater against BA.4/5, BQ.1.1, XBB, and XBB.1.5 (P = .03, .09, .06, and .07, respectively) than titers in individuals who received 4 doses of ancestral mRNA vaccine (V4) (Figure 1B and Supplementary Table 4A). We also compared neutralizing antibody titers against individual variants between each of the antigen exposure groups (V4 vs V3 + Bi vs V3 + PVI vs V4 + Bi). In this multiple comparison analysis, there were no significant differences for D614G, BA.4/5, BQ.1.1, XBB, or XBB.1.5 (Supplementary Table 4B). Within each V4, V3 + Bi, V3 + PVI, and V4 + Bi group, only D614G versus BA.4/5, BQ.1.1, XBB, and XBB.1.5 had significant differences (all P < .05) (Supplementary Table 4C). Overall, GMTs against D614G were similar among the groups (Figure 1B and 1C, Supplementary Table 4A). Notably, GMTs against BQ.1.1, XBB, and XBB.1.5 were low at levels 3- to 19-fold lower than GMTs against BA.4/5. GMTs against XBB and XBB.1.5 were similar within each exposure group (Figure 1B and 1C, Supplementary Table 4A), but GMTs against XBB.1.5 were even lower. GMTs in the V3 + Bi and V3 + PVI groups were similar, while the V4 + Bi group had modestly lower GMTs (Figure 1B and 1D, Supplementary Table 4A).

DISCUSSION

Although concerns about immune imprinting from prior ancestral vaccine exposures remain [6], we found that a booster dose of bivalent vaccine (V3 + Bi) elicited modestly higher neutralizing titers against most recent Omicron variants, including XBB and XBB.1.5 variants, than 4 doses of ancestral vaccine (V4). Within any antigen-exposure group, titers against XBB and XBB.1.5 were similar, though considerably lower than against BA.4/5, indicating significant antigenic divergence. The neutralizing titers for XBB and XBB.1.5 were very low, with about 30% of subjects below 100. Recent studies suggest that neutralization titers correlate with protection against SARS-CoV-2 infection [7–9]. Persons with extremely low neutralizing titers against XBB and XBB.1.5 may be vulnerable to infection by these strains. However, vaccines with antigens that do not match circulating variants may still provide some degree of protection against severe disease or death. Our neutralizing antibody results align with clinical data showing increased vaccine effectiveness of bivalent boosting against BA.5- and XBB/XBB.1.5-related infections compared to monovalent boosting [10, 11]. As most PVIs occurred when Omicron variants predominated, GMTs from the V3 + Bi, V3 + PVI, and V4 + Bi groups (Figure 1B and 1D) suggest that boosting by an infection or vaccine that more closely matches circulating variants may be better than boosting by the ancestral variant. The modest increase in titers against BA.4/5 after the bivalent booster or Omicron infection may represent a priming response to novel epitopes in the BA.4/5 spike that are not present in the ancestral spike or may be due to affinity maturation to shared epitopes. The current bivalent booster vaccine investigated within this study contains mRNAs encoding a half-dose each of the ancestral and BA.4/5 spike proteins. Whether a first monovalent booster with a full dose of an Omicron variant, a second bivalent booster, or a subsequent Omicron exposure would improve titers to Omicron variants remains to be determined. Nonetheless, because the bivalent vaccine and PVI groups elicited strong boosting to the prior D614G variant, evaluation of booster vaccines including only new variants appears warranted. Furthermore, to improve protection against variants with high antigenic divergence from the vaccine antigens, such as XBB and XBB.1.5, updated COVID-19 vaccines containing matched antigens to divergent variants are needed. Study limitations include small sample numbers, a generally healthy cohort with few comorbidities, modestly different median ages among the groups, potential asymptomatic infections, and noncontemporaneous sample collections. Study strengths include well-characterized vaccinees followed prospectively and a broad number of Omicron variants tested.

Supplementary Data

Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Author contributions. Concept and design: W. W., S. P., E. M., and C. D. W. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: W. W. and C. D. W. Critical revision of the manuscript for important intellectual content: All authors. Statistical analysis: W. W. Obtained funding: E. M. and C. D. W. Administrative, technical, or material support: S. P. P. and R. V. Supervision: S. P., E. M., and C. D. W.

Acknowledgments. The authors gratefully acknowledge all research volunteers for their time and participation and all PASS study team members for their contributions to the PASS study.

Data sharing. Data are available in Supplementary Tables 3 and 4. Questions or additional requests should be directed to the corresponding author.

Disclaimer. The sponsors had no involvement in the study design, the collection of data, the analysis or interpretation of data, the writing of the report, or the decision to submit the article for publication. The contents of this publication are the sole responsibility of the author(s) and do not necessarily reflect the views, opinions, or policies of the US Food and Drug Administration (FDA); the Uniformed Services University of the Health Sciences (USU); the Department of Defense (DoD); the Departments of the Army, Navy, or Air Force; the Defense Health Agency; Walter Reed National Military Medical Center (WRNMMC); or the Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc (HJF). Mention of trade names, commercial products, or organizations does not imply endorsement by the US government. The investigators have adhered to the policies for protection of human subjects as prescribed in 45 Code of Federal Regulations 46. R. V., W. W., E. M., and C. D. W. are employees of the US government; this work was prepared as part of their official duties. Title 17 USC §105 provides that copyright protection under this title is not available for any work of the US government. Title 17 USC §101 defines a US government work as a work prepared by a military service member or employee of the US government as part of that person's official duties. E. G. and S. P. are employed by HJF.

Financial support. The protocol was executed by the Infectious Disease Clinical Research Program (IDCRP), a DoD program executed by USU through a cooperative agreement with HJF. This work was supported in whole, or in part, with federal funds from the US FDA Medical Countermeasures Initiative (grant number OCET 2022-1750 to C. D. W.), the Defense Health Program (award numbers HU00012020067 and HU00012120094 to E. M.), the Immunization Healthcare Branch of the Defense Health Agency, US DoD (award number HU00012120104 to E. M.), and the National Institute of Allergy and Infectious Diseases (award number IDCRP HU0001920111 to S.P.) under Inter-Agency Agreement Y1-AI-5072.