Immune escape is observed with SARS-CoV-2 Omicron (Pango lineage B.1.1.529), the predominant circulating strain worldwide. A booster dose was shown to restore immunity against Omicron infection; however, real-world data comparing mRNA (BNT162b2; Comirnaty) and inactivated vaccines’ (CoronaVac; Sinovac) homologous and heterologous boosting are lacking. A retrospective study was performed to compare the rate and outcome of COVID-19 in healthcare workers (HCWs) with various vaccination regimes during a territory-wide Omicron BA.2.2 outbreak in Hong Kong. During the study period from 1 February to 31 March 2022, 3167 HCWs were recruited, and 871 HCWs reported 746 and 183 episodes of significant household and non-household close contact. A total of 737 HCWs acquired COVID-19, all cases of which were all clinically mild. Time-dependent Cox regression showed that, compared with two-dose vaccination, three-dose vaccination reduced infection risk by 31.7% and 89.3% in household contact and non-household close contact, respectively. Using two-dose BNT162b2 as reference, two-dose CoronaVac recipient had significantly higher risk of being infected (HR 1.69 p pp = 0.0157) were associated with a lower risk of infection. Three-dose CoronaVac and two-dose BNT162b2 + CoronaVac booster were not significantly different from two-dose BNT162b2. The mean time to achieve negative RT-PCR or E gene cycle threshold 31 or above was not affected by age, number of vaccine doses taken, vaccine type, and timing of the last dose. In summary, we have demonstrated a lower risk of breakthrough SARS-CoV-2 infection in HCWs given BNT162b2 as a booster after two doses of BNT162b2 or CoronaVac.
