Estimation of the correlates of protection of the rVSVΔG-ZEBOV-GP Zaire ebolavirus vaccine: a post-hoc analysis of data from phase 2/3 clinical trials.

Background: Establishment of immune correlates of protection can provide a measurable criterion for assessing protection against infection or disease. For some vaccines, such as the measles vaccine, antibodies serve as the correlate of protection, but for others, such as human papillomavirus, the correlate of protection remains unknown. Merck & Co, Kenilworth, NJ, USA, in collaboration with multiple partners, developed a live recombinant vesicular stomatitis virus vaccine (rVSVΔG-ZEBOV-GP [ERVEBO]) containing the Zaire ebolavirus glycoprotein (GP) in place of the recombinant vesicular stomatitis virus GP to prevent Ebola virus disease. Seroresponse, defined as post-vaccination GP-ELISA of 200 ELISA units (EU) per mL or higher and two-times or more above baseline, was proposed; however, correlates of protection have not been determined. The objective of this post-hoc analysis was to infer possible correlates of protection for rVSVΔG-ZEBOV-GP.
Methods: In this post-hoc analysis we included vaccinated participants with serology data from three phase 2/3 immunogenicity trials in Guinea, Sierra Leone, and Liberia (n=2199). Two of the trials were open-label, single-arm trials (one randomised [STRIVE], one non-randomised [FLW]); and one trial was randomised, placebo-controlled with two vaccine comparators (PREVAIL). Endpoints were total IgG antibody response (EU per mL) to rVSVΔG-ZEBOV-GP measured by GP-ELISA and neutralising antibody response to rVSVΔG-ZEBOV-GP measured by plaque reduction neutralisation test at days 14, 28, 180, and 365 after vaccination. Reverse cumulative distribution curves of the antibody concentrations were used to estimate statistical correlates of protection between 70% and 100% that might be applied to vaccine efficacy and effectiveness estimates.
Findings: Although GP-ELISA and plaque reduction neutralisation tests showed similar response patterns, GP-ELISA provided a wider range of measurable titres and better differentiation for estimating correlates of protection compared with the plaque reduction neutralisation test. At day 14 after vaccination in the FLW trial, 1060 (100%) of 1060 participants had GP-ELISA levels at or above 68 EU per mL and 742 (70%) of 1060 had levels at or above 313 EU per mL. At day 28 after vaccination in the pooled population, 1953 (100%) of 1953 participants had levels at or above 73 EU per mL and 1368 (70%) of 1953 participants had levels at or above 735 EU per mL. GP-ELISA seroresponse 200 EU per mL or higher and two-times or more increase in antibody level from baseline occurred in 80% or higher of participants at each assessment and in 94% or higher of participants at any time after vaccination.
Interpretation: Our results are consistent with previous work suggesting that seroresponse defined as GP-ELISA of 200 EU per mL or higher and two-times or more from baseline associated with vaccination might be the most appropriate dichotomous correlate of protection and falls within the seroprotective threshold range described herein.
Funding: Merck Sharp & Dohme, Biomedical Advanced Research and Development Authority, Office of the Assistant Secretary for Preparedness and Response, US Department of Health and Human Services.
(Copyright © 2021 The Author(s). Published by Elsevier Ltd. This is an Open Access article under the CC BY 4.0 license. Published by Elsevier Ltd.. All rights reserved.)