Effect of Omicron BA.1-based compared to prototype booster mRNA vaccination on incidence of COVID-19 in the COVAIL trial

Abstract

Covid-19 vaccines are updated to match circulating strains based on reasoning that better strain-matched immunogenicity should provide better protection. Randomized evidence with disease endpoints to support strain matching is lacking. We evaluated COVID-19 incidence among adults randomized to a second booster of Prototype or Omicron-based vaccines. COVAIL was a four-stage Phase 2 clinical trial; results from Stages 1 (mRNA-1273 [Moderna]) and 2 (BNT162b2 [Pfizer/BioNTech]) are described here. Adults who had received a primary series and one booster of an authorized COVID-19 vaccine were eligible. Participants received one dose of either Prototype vaccine or a monovalent or bivalent Omicron BA.1 vaccine. SARS-CoV-2 neutralization titers (ID50) were measured pre- and post-vaccination. Covariate-adjusted cumulative COVID-19 incidence and Cox regression analyses were conducted separately for each stage. 706 participants with pre- and day 15 post-vaccination ID50 titers (n = 503 in Stage 1, n = 203 in Stage 2) were included. Within stages, participant characteristics and baseline ID50 titers were similar between Prototype and Omicron-based arms. There was no difference in cumulative COVID-19 incidence for Prototype vs. Omicron-based vaccine in Stage 1 (RR 1.04, 95 % CI 0.73–1.48), while incidence was higher among Prototype recipients in Stage 2 (RR 2.56, 1.44–4.52). Cox regression analysis showed no difference in Stage 1 (HR 1.04, 0.68–1.58), but higher incidence for Prototype recipients in Stage 2 (HR 2.95, 1.52–5.72). Omicron-based vaccines as second boosters were more protective against COVID-19 relative to Prototype among those receiving BNT162b2 but not mRNA-1273. Differences between stages such as force of infection, antigen matching, and vaccine differences may explain this finding.

Bo Zhang

Assistant Professor of Biostatistics

My research interests include design of observational studies, instrumental variables, application of causal inference in medicine and applied statistics in general.