Bivalent Omicron BA.1-Adapted BNT162b2 Booster in Adults Older than 55 Years

Abstract

Background: The emergence of immune-escape variants of severe acute respiratory syndrome coronavirus 2 warrants the use of sequence-adapted vaccines to provide protection against coronavirus disease 2019.

Methods: In an ongoing phase 3 trial, adults older than 55 years who had previously received three 30-μg doses of the BNT162b2 vaccine were randomly assigned to receive 30 μg or 60 μg of BNT162b2, 30 μg or 60 μg of monovalent B.1.1.529 (omicron) BA.1-adapted BNT162b2 (monovalent BA.1), or 30 μg (15 μg of BNT162b2 + 15 μg of monovalent BA.1) or 60 μg (30 μg of BNT162b2 + 30 μg of monovalent BA.1) of BA.1-adapted BNT162b2 (bivalent BA.1). Primary objectives were to determine superiority (with respect to 50% neutralizing titer [NT₅₀] against BA.1) and noninferiority (with respect to seroresponse) of the BA.1-adapted vaccines to BNT162b2 (30 μg). A secondary objective was to determine noninferiority of bivalent BA.1 to BNT162b2 (30 μg) with respect to neutralizing activity against the ancestral strain. Exploratory analyses assessed immune responses against omicron BA.4, BA.5, and BA.2.75 subvariants.

Results: A total of 1846 participants underwent randomization. At 1 month after vaccination, bivalent BA.1 (30 μg and 60 μg) and monovalent BA.1 (60 μg) showed neutralizing activity against BA.1 superior to that of BNT162b2 (30 μg), with NT₅₀ geometric mean ratios (GMRs) of 1.56 (95% confidence interval [CI], 1.17 to 2.08), 1.97 (95% CI, 1.45 to 2.68), and 3.15 (95% CI, 2.38 to 4.16), respectively. Bivalent BA.1 (both doses) and monovalent BA.1 (60 μg) were also noninferior to BNT162b2 (30 μg) with respect to seroresponse against BA.1; between-group differences ranged from 10.9 to 29.1 percentage points. Bivalent BA.1 (either dose) was noninferior to BNT162b2 (30 μg) with respect to neutralizing activity against the ancestral strain, with NT₅₀ GMRs of 0.99 (95% CI, 0.82 to 1.20) and 1.30 (95% CI, 1.07 to 1.58), respectively. BA.4-BA.5 and BA.2.75 neutralizing titers were numerically higher with 30-μg bivalent BA.1 than with 30-μg BNT162b2. The safety profile of either dose of monovalent or bivalent BA.1 was similar to that of BNT162b2 (30 μg). Adverse events were more common in the 30-μg monovalent-BA.1 (8.5%) and 60-μg bivalent-BA.1 (10.4%) groups than in the other groups (3.6 to 6.6%).

Conclusions: The candidate monovalent or bivalent omicron BA.1-adapted vaccines had a safety profile similar to that of BNT162b2 (30 μg), induced substantial neutralizing responses against ancestral and omicron BA.1 strains, and, to a lesser extent, neutralized BA.4, BA.5, and BA.2.75 strains. (Funded by BioNTech and Pfizer; ClinicalTrials.gov number, NCT04955626.).

Figure 1. Randomization and Follow-up.

Shown is the randomization and follow-up of participants in the expanded cohort as of the database cutoff date of May 16, 2022. Adults older than 55 years of age who had previously received three 30-μg doses of the BNT162b2 vaccine were randomly assigned in a 1:1:1:1:1:1 ratio to receive the original 30-μg dose or a 60-μg dose of BNT162b2, a 30-μg or 60-μg dose of monovalent B.1.1.529 (omicron) BA.1–adapted BNT162b2 vaccine (monovalent BA.1), or bivalent combinations thereof (30-μg bivalent BA.1 [15 μg of BNT162b2+15 μg of monovalent BA.1] or 60-μg bivalent BA.1 [30 μg of BNT162b2+30 μg of monovalent BA.1]). At the data-cutoff date, some participants had not reached the 1-month postvaccination visit. Participants were recruited from 36 sites in the United States. Randomization and follow-up in the sentinel cohort are shown in Figure S1 in the Supplementary Appendix.

Figure 2. SARS-CoV-2 Neutralization Assay Results at 1 Month after Vaccination among Participants without Evidence of Previous SARS-CoV-2 Infection.

Shown are the geometric mean titers (GMTs) of neutralizing antibodies against the omicron BA.1 strain (Panel A) and the ancestral USA-WA1/2020 strain (Panel B) before and 1 month after vaccination and the associated geometric mean fold rises from baseline. The analyses were performed in the immunogenicity subset of 230 participants who had been randomly selected from each treatment group in the expanded cohort to evaluate primary and secondary immunogenicity objectives, and the results are shown for those who had immunogenicity data that could be evaluated (according to the definitions in Table S1). Before the 1-month postvaccination blood sample collection, participants had no serologic or virologic evidence of previous severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (i.e., a negative SARS-CoV-2 nucleoprotein–binding antibody test at the trial vaccination visit and the 1-month postvaccination visit and a negative nucleic acid amplification test at the trial vaccination visit and any unscheduled visit before the 1-month postvaccination blood sample collection) and no medical history of Covid-19. The GMTs are shown within the bars (𝙸 bars indicate the 95% confidence interval), and the geometric mean fold rises from before to 1 month after vaccination are shown above the bars. The GMTs, geometric mean fold rises, and associated 95% confidence intervals were calculated by exponentiating the mean logarithm of the 50% neutralizing titers or fold rises and the corresponding 95% confidence intervals on the basis of the Student’s t-test distribution; assay results below the lower limit of quantitation (LLOQ) were imputed as 0.5×LLOQ (the LLOQ is 32 for the neutralizing titer against BA.1 and 87 for the neutralizing titer against the ancestral strain). The 30-μg dose of BNT162b2 was the original dose the participants had received at three previous vaccinations. The 30-μg dose of bivalent BA.1 contained 15 μg of BNT162b2plus15 μg of monovalent BA.1, and the 60-μg dose contained 30 μg of BNT162b2 plus 30 μg of monovalent BA.1.

Figure 3. SARS-CoV-2 Fluorescent Focus Reduction Neutralization Test Results for Omicron Subvariants.

Panel A shows the GMTs against omicron BA.4 and BA.5 subvariants and the geometric mean fold rises from before to 1 month after vaccination among participants in the omicron BA.4–BA.5 neutralization assay subset, which included 100 participants (20 with SARS-CoV-2–positive status at baseline and 80 with SARS-CoV-2–negative status up to 1 month after vaccination) who had been randomly selected from each of the 30-μg BNT162b2 and 30-μg bivalent-BA.1 groups in the expanded cohort. Panel B shows GMTs against the omicron BA.2.75 subvariant 1 month after vaccination among participants in the omicron BA.2.75 neutralization assay subset, which included 30 participants in each of the 30-μg BNT162b2 and 30-μg bivalent-BA.1 groups who had been randomly selected from the omicron BA.4–BA.5 neutralization assay subset, all of whom did not have serologic or virologic evidence of previous SARS-CoV-2 infection up to 1 month after vaccination. SARS-CoV-2–positive status was defined as a positive SARS-CoV-2 nucleoprotein–binding antibody test at baseline, a positive nucleic acid amplification test at baseline, or a medical history of Covid-19. The GMTs are shown within the bars in Panels A and B (𝙸 bars indicate the 95% confidence interval), and the geometric mean fold rises from before to 1 month after vaccination are shown above the bars in Panel A. The GMTs, geometric mean fold rises, and associated 95% confidence intervals were calculated by exponentiating the mean logarithm of the 50% neutralizing titers or fold rises and the corresponding 95% confidence intervals on the basis of the Student’s t distribution. Assay results below the LLOQ were imputed as 0.5×LLOQ (the LLOQ for the neutralizing titers against BA.4, BA.5, and BA.2.75 by the fluorescent focus reduction neutralization test is 20).