Relative effectiveness and durability of booster doses of SARS-CoV-2 vaccines: A systematic review and meta-analysis

Abstract

<p>Billions of people worldwide have received booster doses of SARS-CoV-2 vaccines. Continuous monitoring of the relative vaccine effectiveness (rVE) and durability of booster vaccination over previous vaccinations is important for developing vaccination strategies during the post-pandemic era. We conducted a systematic review and meta-analysis of trials and observational studies to determine the rVE of the 1^st booster over no booster and that of the 2^nd booster over the 1^st booster. Three trials and 91 observational studies were included in this systematic review. For the 1^st booster homologous vaccination, the rVE at a median of 9 weeks after booster vaccination was 66.9% (95% confidence interval: 59.8%, 72.7%), 75.9% (62.6%, 84.5%), 74.1% (66.9%, 79.8%), 86.1% (78.7%, 90.9%) and 84.2% (78.3%, 88.5%) against overall infection, symptomatic infection, hospital admission, severe infection and COVID-19-related death, respectively. The rVE against overall infection was affected by variants of concern and vaccine types and waned by average of 4.3% (3.3%, 5.4%; <i>P-trend</i><0.01) per week. Heterologous regimens for the 1^st booster vaccination demonstrated effectiveness comparable to that of homologous regimens. The rVE of the 2^nd booster homologous vaccination at a median of 7 weeks after booster vaccination was 41.9% (31.2%, 51.0%), 53.1% (24.5%, 70.9%), 60.6% (55.3%, 65.3%), 56.4% (45.3%, 65.2%) and 68.2% (51.2%, 79.2%) against the five outcomes above, respectively, with no significant decrease in the rVE of 2^nd booster vaccination. In conclusion, both the 1^st and 2^nd booster homologous vaccinations provided additional protection against mild and severe infections. The rVE of the 1^st booster rapidly waned over time. The rVE of the 2^nd booster, including heterologous vaccination, its durability and the sources of heterogeneity, however, remains uncertain and more relevant studies are needed.</p>