Second booster doses of adenoviral- and mRNA-based COVID-19 vaccines increase protection against COVID-19 hospitalization: Final analysis from the REFORCO-Brazil real-world effectiveness study during Omicron.

Background: Booster doses of COVID-19 vaccines are required to maintain protection against SARS-CoV-2. However, real-world evidence from South America, needed to inform optimal vaccination strategies, is lacking. Herein, we present the final analysis of REFORCO-Brazil, a large-scale assessment of relative vaccine effectiveness (rVE) of second boosters (vs first boosters) against hospitalization with COVID-19.

Methods: REFORCO-Brazil is a test-negative case-control study (NCT05697705) that utilized Brazilian national data on severe acute respiratory syndrome (SARS) surveillance and COVID-19 vaccination. Individuals hospitalized with SARS from January 1 to December 31, 2022, were classified as test-positive cases (via SARS-CoV-2 antigen/reverse transcription polymerase chain reaction [RT-PCR]) or test-negative case-controls (via RT-PCR) and matched by admission date, region, sex, preceding COVID-19 vaccinations, and age. We used conditional logistic regression combined with multiple covariate adjustments to estimate rVE for second boosters (versus first boosters received ≥4 months prior) overall, by type (AZD1222, Ad26.COV2·S, BNT162b2, and CoronaVac) and in vulnerable subgroups (elderly and immunocompromised/high-risk individuals).

Results: Median (range) time between second booster and SARS hospitalization was 87.0 (8.0-307.0) and 79.0 (8.0-303.0) days among 5426 test-positive cases and 6131 test-negative controls, respectively. Overall rVE of any second booster against hospitalization was 18.7 % (95 % confidence interval [CI]: 10.5-26.1). The rVE of adenoviral- and mRNA-based vaccines was similar; 18.2 % (4.8-29.8) for AZD1222, 20.7 % (10.2-30.0) for Ad26.COV2·S, and 23.2 % (9.7-34.7) for BNT162b2. Similar levels of added protective benefit, or 'boosting', was observed in very elderly and immunocompromised/high-risk individuals. Additional protection was highest within 2 months post-dosing, decreasing thereafter. Exploratory analyses revealed increased protection against severe in-hospital outcomes, including mortality.

Conclusions: Our results support the use of monovalent adenoviral/mRNA-based vaccine maintain protection against COVID-19 hospitalization from Omicron subvariants. However, optimal timing of booster vaccinations will need to be carefully considered for future booster strategies, especially among vulnerable subgroups.