Effectiveness of BA.1-and BA.4/BA. 5-Containing Bivalent COVID-19 mRNA Vaccines Against Symptomatic SARS-CoV-2 Infection During the BA.5-Dominant Period in Japan

The initial high efficacy and effectiveness of mRNA vaccines against COVID-19 waned due to the emergence of immune-escape variants. Bivalent vaccines, containing mRNA for the ancestral strain and either Omicron subvariant BA.1 or BA.4/BA.5, were developed to address this. Japan approved both BA.1-containing and BA.4/BA.5-containing bivalent vaccines in late 2022. However, real-world data on their effectiveness was lacking. Japan's population, with a significant proportion being infection-naive, provided a unique opportunity to assess vaccine effectiveness (VE). This study reports the results of a multicenter, prospective, test-negative, case-control study conducted in Japan to evaluate the effectiveness of these bivalent vaccines against symptomatic SARS-CoV-2 infection during the BA.5-dominant period. The study aimed to provide crucial real-world epidemiological data to complement in vitro and animal model data used for initial vaccine approval, addressing the urgent need for such information to guide public health strategies.

The authors cite several studies highlighting the waning immunity of initial COVID-19 mRNA vaccines and the emergence of immune-escape variants, creating the need for updated vaccines. They also mention previous research conducted in Japan on COVID-19 vaccine effectiveness during the Delta and Omicron dominant periods. This literature review sets the stage for the study's rationale, emphasizing the importance of evaluating the real-world effectiveness of the bivalent vaccines in a population with a unique epidemiological profile.

This multicenter, prospective, test-negative, case-control study, the FASCINATE study, involved individuals who visited one of ten healthcare facilities in the Kanto region of Japan between September 20 and December 31, 2022 (BA.5 dominant period). The inclusion criteria were individuals aged ≥16 years, excluding those who could not consent, required immediate treatment, or had previously participated. The study used a test-negative design, comparing individuals with COVID-19-like symptoms who tested positive for SARS-CoV-2 (cases) to those who tested negative (controls). A questionnaire was administered before test results to minimize bias. Vaccination status was recorded based on self-reported vaccine records/certificates. Logistic regression was used to estimate the odds of being vaccinated among cases relative to controls, adjusted for age, sex, comorbidities, occupation, prior SARS-CoV-2 testing/infection, close contact history, healthcare facility, calendar week, mask wearing, high-risk behaviors, and influenza vaccination status. Vaccine effectiveness (VE) was calculated as VE = (1 − adjusted odds ratio [aOR]) × 100%. Both absolute VE (aVE) and relative VE (rVE) were calculated. A head-to-head comparison of bivalent versus monovalent vaccines was performed, accounting for the time elapsed since the last vaccination.

A total of 6955 individuals were enrolled; after exclusions, 6191 were included in the analysis. The study found a high aVE of BA.1-containing bivalent vaccines (65%) and BA.4/BA.5-containing bivalent vaccines (76%) against symptomatic infection compared to no vaccination. However, rVE compared to monovalent vaccines administered more than six months prior was moderate (46%). The head-to-head comparison of bivalent versus monovalent vaccines within a short timeframe after administration did not show bivalent vaccine superiority (aOR, 0.99). This could be due to the unavailability of monovalent vaccines after the introduction of the bivalent vaccine. The observed aVE, while high, was lower than that observed for the monovalent primary series against earlier variants (85%-95%), suggesting potential immune imprinting.

The high aVE of bivalent vaccines against symptomatic infection, compared to no vaccination, supports their continued rollout. The moderate rVE compared to monovalent vaccines suggests added benefit, particularly with a longer interval since the last monovalent dose. The lack of bivalent superiority in the head-to-head comparison could be influenced by factors like the differences in the proportion of previously infected individuals and the differing availability of monovalent and bivalent vaccines. The lower aVE than observed with earlier monovalent vaccines is consistent with the concept of immune imprinting, where prior immunity to the ancestral strain might influence the response to the bivalent vaccine. This study contributes valuable real-world data on bivalent vaccine effectiveness, particularly in a population with a unique epidemiological setting.

This study demonstrated high absolute vaccine effectiveness of both BA.1 and BA.4/BA.5 containing bivalent COVID-19 mRNA vaccines against symptomatic SARS-CoV-2 infection during a BA.5 dominant period in Japan, relative to no vaccination. The findings support the continued use of bivalent vaccines, although further monitoring of mid- to long-term effectiveness is needed. Future research should explore the impact of immune imprinting and the long-term effectiveness of these vaccines in various populations and under different epidemiological conditions.

The study acknowledges inherent biases in observational studies, despite efforts to minimize them through adjustment for various factors. Reliance on self-reported vaccination data introduces potential inaccuracies. The wide confidence intervals for some estimates necessitate cautious interpretation of point estimates. Finally, the analysis was a complete case analysis, excluding individuals with missing data.