Neutralizing antibody responses in healthcare workers vaccinated against COVID-19 in Brazil: impact of primary regimens and an mRNA booster

Background Inactivated COVID-19 vaccines were widely used in Brazil, particularly among healthcare workers. However, the durability of neutralizing antibody responses after primary vaccination, particularly against variants of concern such as Gamma, remained uncertain and contributed to the adoption of booster-dose strategies. This study aimed to evaluate SARS-CoV-2–specific neutralizing antibody responses in healthcare workers who received COVID-19 vaccines, before and after an mRNA booster dose. Methods In this prospective longitudinal study, 117 healthcare workers from a tertiary hospital in São Paulo, Brazil, were enrolled; primary-regimen analyses focused on the 115 participants who had received CoronaVac or ChAdOx1-S. Seventy participants returned for a second blood sample after booster vaccination; the paired BNT162b2 analysis included 67 participants who had received CoronaVac or ChAdOx1-S as the primary series and BNT162b2 as the booster. Neutralizing antibodies were measured using a cytopathic effect-based virus neutralization assay (CPE-VNT100) against the ancestral SARS-CoV-2 strain (Wuhan-Hu-1) and the Alpha, Gamma, and Delta variants. Titers <20 were assigned a value of 10 for GMT calculations, and multiplicity across the four variant-specific comparisons was controlled using the Holm procedure. Results Among CoronaVac recipients without documented prior infection, 70% lacked neutralizing antibodies against the ancestral strain, and up to 80% had no detectable neutralizing activity against the Gamma variant at approximately seven months after primary vaccination. Four of 63 participants with documented infection before booster (6.3%, all CoronaVac recipients) lacked detectable activity against Gamma. Observed pre-booster GMTs were higher among ChAdOx1-S recipients than among CoronaVac recipients for all four viruses (Holm-adjusted p ≤ 0.006), although sampling intervals differed substantially by platform. In the paired BNT162b2 analysis, GMTs increased 20.9-fold against Wuhan-Hu-1, 26.3-fold against Gamma, 17.7-fold against Alpha, and 15.9-fold against Delta (all Holm-adjusted p < 0.001). Excluding the seven participants with documented infection between booster and post-booster sampling yielded similarly large increases (13.4- to 22.9-fold). Only one participant remained seronegative against all four viruses after booster. Conclusions In this cohort, neutralizing activity measured several months after primary vaccination was limited among infection-naïve CoronaVac recipients, particularly against Gamma. Higher pre-booster titers were observed among ChAdOx1-S recipients, although this comparison is confounded by markedly different sampling intervals. A BNT162b2 booster induced a strong and broadly reactive neutralizing response, and the findings were robust after excluding post-booster SARS-CoV-2 infections.