Fig. 6: Vaccine efficacy (solid lines) in baseline SARS-CoV-2 negative per-protocol participants by post-vaccination antibody marker level in four randomized, placebo-controlled COVID-19 vaccine efficacy trials.

Vaccine efficacy (VE) was estimated using the marginalized Cox proportional hazards implementation of ref. ⁴⁵. Vaccine efficacy estimates are shown by a anti-spike IgG concentration [D57 in COVE, D29 in ENSEMBLE-United States sites (ENSEMBLE-US), D35 in PREVENT-19, D56 in COV002] or by b pseudovirus (PsV)-nAb ID50 titer (D57 in COVE, D29 in ENSEMBLE-US, D35 in PREVENT-19, D56 in COV002). The dashed lines indicate bootstrap point-wise 95% confidence intervals. The follow-up periods for the VE assessment were: COVE (doses D1, D29), 7 to 100 days post D57; ENSEMBLE-US (one dose, D1), 1 to 53 days post D29; PREVENT-19 (doses D0, D21), 7 to 59 days post D35; COV002 (doses D0, D28; 28 days post D28 until the end of the study period). The histograms are an estimate of antibody marker density in baseline SARS-CoV-2 negative per-protocol vaccine recipients. Curves are plotted over the following antibody marker ranges: a COVE: 2.5th percentile to 97.5th percentile of marker, ENSEMBLE-US: 2.5th percentile to 97.5th percentile, PREVENT-19: 2.5th percentile to 97.5th percentile, COV002: 29 to 899 BAU/ml; b COVE: 10 IU50/ml to 97.5th percentile of marker, ENSEMBLE-US: 2.5th percentile to 97.5th percentile, PREVENT-19: 2.5th percentile to 97.5th percentile, COV002: 3 to 140 IU50/ml. Baseline covariates adjusted for were: COVE: baseline risk score, comorbidity status, and Community of color status; ENSEMBLE-US, baseline risk score; PREVENT-19: baseline risk score; COV002: baseline risk score. Source data are provided in Fig. 6 Source Data file. BAU binding antibody unit; D0, Day 0 visit; D1, Day 1 visit; D35, Day 35 visit; D59, Day 59 visit; ID50, 50% inhibitory dilution; IU50 international unit 50; nAb neutralizing antibody; RBD receptor binding domain; ULoQ upper limit of quantification.