OBJECTIVES: For Ebola vaccine development, antibody response is a major endpoint although its determinants are not well known. We aimed to review Ebola vaccine studies and to assess factors associated with antibody response variability in humans. METHODS: We searched PubMed and Scopus for preventive Ebola vaccine studies in humans or non-human primates (NHP), published up to February 2018. For each vaccination group with Ebola Zaire antibody titre measurements after vaccination, data about antibody response and its potential determinants were extracted. A random-effects meta-regression was conducted including human groups with at least 8 individuals. RESULTS: We reviewed 49 studies (202 vaccination groups including 74 human groups) with various vaccine platforms and antigen inserts. Mean antibody titre was slightly higher in NHP (3.10, 95% confidence interval [2.93; 3.27]) than in humans (2.75 [2.57; 2.93]). Vaccine platform (p<0.001) and viral strain used for antibody detection (p<0.001) were associated with antibody response in humans, but adjusted heterogeneity remained at 95%. CONCLUSIONS: Various platforms have been evaluated in humans, including Ad26, Ad5, ChimpAd3, DNA, MVA, and VSV. In addition to platforms, viral strain used for antibody detection influences antibody response. However, variability remained mostly unexplained. Therefore, comparison of vaccine immunogenicity needs randomised controlled trials.Read less <