Differential prolonged multiomic responses to mRNA and inactivated virus COVID-19 vaccines

Abstract Despite the significant impact of SARS-CoV-2 vaccines in curtailing the spread and severity of COVID-19, comprehensive omics-level understanding of longitudinal host responses remains elusive. We recruited 553 Hong Kong participants receiving two doses of CoronaVac (inactivated virus) or BNT162b2 (mRNA-based). Longitudinal blood samples (baseline, one-month, six-months post-vaccination) were collected. We assessed the longitudinal changes in omics landscape and their impact on host immunity by applying transcriptomic, metabolomic, lipidomic and cytokine datasets. Our data delineate pronounced differences in the multi-omics profiles between the two vaccine platforms examined. The mRNA-based BNT162b2 vaccine induced more profound changes in omics profiles compared to the inactivated CoronaVac, with notable alterations in adaptive immunity and metabolomic functions. Importantly, BNT162b2 displayed a durable immunological imprint persisting at six months, characterized by persistent immune gene modifications and metabolic reprogramming. Findings also reveal greater omics crosstalk in response to BNT162b2, which is related to the enhancing of its immunogenicity and durability. Furthermore, predictive modeling demonstrated that baseline multi-omics profiles forecast vaccine-specific performance (AUC: 0.73–0.79). Collectively, this work identifies vaccine-associated longitudinal multi-omics signatures after primary SARS-CoV-2 vaccination. Our results advocate integrating multi-omics data to tailor vaccination strategies to individual immune profiles, enhancing public health outcomes.