Booster immunization improves immune responses against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants and vaccine effectiveness against coronavirus disease 2019 (COVID-19).1-3 The emergence of antigenically-divergent Omicron variants, required updated booster immunization strategies.4-6 As such, Omicron-containing, bivalent boosters are currently available in multiple geographies to address COVID-19 caused by omicron variants.7,8
We previously reported the day 29 interim results of an ongoing phase 2/3 study (NCT04927065) that evaluates the safety and immunogenicity of the Omicron-BA.1-containing bivalent booster mRNA-1273.214 administered as a second booster dose to participants who have previously received the original mRNA-1273 vaccine.9 The mRNA-1273.214 50-µg booster elicited higher neutralizing antibody (nAb) responses against Omicron BA.1, compared to the 50-µg mRNA-1273 booster, and exhibited a cross-neutralization ability against multiple Omicron variants including BA.4/BA.5 and BA.2.75 with a safety profile similar to that of mRNA-1273.9,10 Although increased potency and breadth of the antibody response are highly desirable, it is also important to evaluate antibody persistence. Here, we present day 91 mRNA-1273.214 immunogenicity and safety data to address the question of antibody durability and longer-term safety with bivalent boosters. The day 91 immunogenicity objectives were pre-specified to evaluate whether the bivalent booster can induce superior responses, compared to mRNA-1273, 3 months after the booster dose.
Between February 18th-March 8th, 2022 (part F, cohort 2) and March 8th-March 23rd, 2022 (part G), 819 participants were enrolled who had previously received the primary series of 100-µg mRNA-1273 and a first booster dose of 50-µg mRNA-1273, ≥3 months prior (Figure S1).9 Participant demographics and baseline characteristics were balanced in the 50-µg mRNA-1273.214 (n=437) and 50-µg mRNA-1273 (n=376) groups (Table S1) including age, race and ethnicity. Median interval days (interquartile range [IQR)]) between second doses of mRNA-1273 in the primary series and the first booster of mRNA-1273 (245 [224-275] and 242 [225-260]) and between the first booster dose of mRNA-1273 and the second booster doses (136 [118-150] and 134 [118-150]) were similar between the mRNA-1273.214 and mRNA-1273 groups. Additionally, similar percentages of participants had evidence of prior SARS-CoV-2-infection in the mRNA-1273.214 (22%) and mRNA-1273 (27%) groups.
Median durations of follow-up days (IQR) were 113 (111-115) for the mRNA-1273.214 and 127 (125-132) for the mRNA-1273 boosters. The occurrences of solicited adverse reactions (ARs) within 7 days and the incidence of unsolicited adverse events (AEs) up to 28 days following mRNA-1273.214 were overall similar to those of mRNA-1273 as previously reported (Tables S2-S3).9 In this interim analysis with longer-term follow-up than previously reported, the incidences of unsolicited AEs reported throughout to the cutoff date regardless of relationship to study vaccine (47.8% and 52.1%) and those considered related to study vaccination by the investigator (4.8% and 5.6%) were similar between the mRNA-1273.214 and the mRNA-1273 groups, respectively (Table S4). One (0.3%) death (fatal hypotension in a 73-year-old female with pre-existing vascular conditions who had an elective cardiac catheterization complicated by hypotension) occurred 64 days after mRNA-1273 immunization and was considered by the investigator to not be related to study vaccination. No events of myocarditis or pericarditis were reported and a total of 8 (1.8%) serious AEs occurred in the mRNA-1273.214 and 10 (2.7%) in the mRNA-1273 groups, all of which were considered unrelated to study vaccination by investigators (Tables S5).
Immunogenicity was evaluated per the pre-specified testing strategy (Figure S2) at days 29 and 91 post-booster in participants in the per-protocol immunogenicity set without evidence of pre-booster SARS-CoV-2-infection (PPIS-negative), as well as in the per-protocol immunogenicity set which includes all participants regardless of pre-booster infection status (PPIS) (Figure S3). Previously published results included only day 29 antibody responses.9 Neutralizing antibody (nAb) geometric mean titers (GMTs) 95% confidence intervals (95% CI) at day 29 post-boost in the PPIS-negative set against Omicron BA.1 were higher after the mRNA-1273.214 (2366.6 [2066.2-2710.7]) than mRNA-1273 (1468.7 [1266.2-1703.6]) booster, and similar against ancestral SARS-CoV-2 (D614G) (5968.1 [5315.4-6700.9]) and (5651.4 [5055.7- 6317.3]) following the boosters, respectively (Table 1 and Figure S4). Estimated GMTs (95% CI) after adjusting for age groups and pre-booster titers at day 29 against Omicron BA.1 were 2469.7 (2255.5-2704.3) and 1419.1 (1280.8-1572.3) and against ancestral SARS-CoV-2 (D614G) were 6406.1 (5975.5-6867.8) and 5291.1 (4890.5-5724.5) after the mRNA-1273.214 and mRNA-1273 boosters, respectively. At day 91, observed GMTs (95% CI) against Omicron BA.1 were also higher after the mRNA-1273.214 (964.4 [834.4-1114.7]) than mRNA-1273 (624.2 [533.1-730.9]) booster and those against ancestral SARS-CoV-2 (D614G) were similar (3428.3 [3062.7-3837.6] and 3447.1 [3054.7-3889.9]) for the boosters. Estimated GMT (95% CI) against Omicron BA.1 were 997.5 (898.4-1107.4) and 602.7 (534.7-679.4) and against ancestral SARS-CoV-2 (D614G) were 3595.6 (3334.8-3876.8) and 3257.3 (2986.3-3552.9) for mRNA1273.214 and mRNA-1273, respectively. The geometric mean ratios (GMRs [97.5% CI]) for the mRNA-1273.214 versus mRNA-1273 GMTs against Omicron BA.1 at both days 29 (1.74 [1.49-2.04]) and 91 (1.66 [1.38-1.99]) met the pre-specified criterion (lower bound of the 97.5% CI of GMR >1) for superiority, and the GMRs against ancestral SARS-CoV-2 (D614G) at both days 29 (1.21 [1.07-1.37]) and 91 (1.10 [0.97-1.26]), met the pre-specified criterion for non-inferiority (lower bound of 97.5% CI of GMR ≥0.67).
Seroresponse rates (SRR [95% CI]) at day 29 were 100% (98.9−100%) and 99.2% (97.2−99.9%) against Omicron BA.1 and 100% (98.9−100% and 98.6−100%) against ancestral SARS-CoV-2 (D614G) for the mRNA-1273.214 and mRNA-1273 boosters, respectively, and at day 91, SRRs were 98.5% (96.4−99.5%) and 96.3% (97.2−99.9%) against Omicron BA.1 and 100% (98.9−100%) and 99.6% (97.7−100.0%) against ancestral SARS-CoV-2 (D614G). The estimated SRR (97.5% CI) differences between mRNA-1273.214 and mRNA-1273 at days 29 (1.5% [-1.1−4.1%] and 0) and 91 (2.1% [-1.6-5.8%] and 0.9% [-1.6−3.5%]) against Omicron BA.1 and ancestral SARS-CoV-2 (D614G), respectively, both met the prespecified non-inferiority criterion (lower bound of the 97.5% CI >-10%).
Results were consistent in all participants regardless of prior SARS-CoV-2 infection (PPIS) and the PPIS participants who had evidence of pre-booster SARS-CoV-2 infection, with higher nAb titers observed at days 29 and 91 against Omicron BA.1 for the mRNA-1273.214 versus mRNA-1273 booster and similar titers against ancestral SARS-CoV-2 (D614G) at both days for the two boosters (Figure 1 and Table S6). Neutralizing titers against Omicron BA.1 were also consistently higher with mRNA-1273.214 than mRNA-1273 at days 29 and 91 among those ≥65 years and 18-<65 years of age in the PPIS-negative group (Figure S5 and Table S7). The GMTs against ancestral SARS-CoV-2 (D614G) were similar at day 29 and day 91 among those ≥65 years and 18-<65 years respectively in the PPIS-negative set.
In participants in the PPIS set with no prior SARS-CoV-2-infection, binding antibody (bAb) GM-levels were higher for mRNA-1273.214 than mRNA-1273 against both Omicron BA.1 and ancestral SARS-CoV-2 (D614G), and alpha, gamma, and delta variants at both days 29 and 91 (Figure S6 and Tables S8-S9). Across variants, the GMRs of bAb levels for mRNA-1273.214 versus mRNA-1273 ranged from 1.10 (1.03-1.19) to 1.24 (1.15-1.34) at day 29 and 1.20 (1.11-1.30) to 1.26 (1.16-1.37) at day 91. Seroresponse rates of 100% were observed for all variants and the differences were 0.
As of the data cutoff date (July 6, 2022), among all participants regardless of pre-booster SARS-CoV-2-infection status, the incidences of SARS-CoV-2-infection and COVID-19 events starting 14 days post-booster were balanced between the mRNA-1273.214 and mRNA-1273 groups (Table S10). There were 61 (14.0%) and 48 (12.8%) SARS-CoV-2-infections, 19 (4.3%) and 18 (4.8%) asymptomatic infections, and 39 (8.9%) and 26 (6.9%) COVID-19 events per the COVE trial definition, and 42 (9.6%) and 30 (8.0%) per the CDC definition in the mRNA-1273.214 and mRNA-1273 groups, respectively. The incidences of infections were similar in the per-protocol efficacy set of participants with no prior SARS-CoV-2-infection. No emergency room visits, or hospitalizations attributed to COVID-19 were seen.
Limitations of the study include that it was not randomized and although a sequential design was used, baseline characteristics, including intervals between prior doses, were balanced between the two groups. Although the incidence of infections post-booster is provided as an exploratory analysis, the study was not designed to evaluate vaccine efficacy; post-booster effectiveness will need to be evaluated in observational studies. Only humoral responses were assessed and cellular responses warrant characterization in ongoing studies.11
In conclusion, the Omicron BA.1-containing bivalent booster vaccine mRNA-1273.214 induced more durable neutralizing antibody responses against BA.1 compared to the original mRNA-1273 with a reassuring safety profile through approximately 3.5 months after the bivalent booster dose. Neutralizing antibody titers against Omicron BA.1 were significantly higher 90 days after the booster dose with the bivalent than the mRNA-1273 vaccine, with no decrement in the response against ancestral SARS-CoV-2 (D614G). The binding antibody data also indicate improved antibody persistence with the bivalent booster against multiple variants. The higher antibody titers at days 29 and 91 with the bivalent booster could, at least in part, be due to new, variant-specific germinal centers induced after immunization with variant-targeting boosters.11 Neutralizing antibody responses were generally consistent regardless of age and pre-booster SARS-CoV-2-infection, and the binding antibody data indicate improved antibody persistence against multiple variants. Although enhanced antibody responses have the potential to confer improved protection against COVID-19, real-world vaccine effectiveness studies are needed to address this question.