Characteristics of the Participants
Characteristics of the Participants at Baseline.
From April through December 2018, a total of 1400 adults and 1401 children were enrolled in the two trials (Figs. S1 and S2). The characteristics of the participants in the randomized groups were balanced at baseline (Table 1).
In the trial involving adults, the median age was 27 years (interquartile range, 20 to 38), and 45% of the participants were women. At baseline, 15% of the adults had an antibody concentration of at least 200 EU per milliliter. The percentage of participants with an antibody concentration of at least 200 EU per milliliter at baseline was higher in Guinea (13%), Liberia (21%), and Sierra Leone (21%) than in Mali (5%) (Table S1).
In the trial involving children, one third of the participants were enrolled in each of the three age groups: 1 to 4 years, 5 to 11 years, and 12 to 17 years; 46% of the children were female. At baseline, 12% of the children had an antibody concentration of at least 200 EU per milliliter. As was observed with the adults, among children 12 to 17 years and 5 to 11 years of age, the percentage of participants with an antibody concentration of at least 200 EU per milliliter was lower in Mali than other countries; this relationship was not seen among participants 1 to 4 years of age.
Adherence to Booster Vaccination and Completeness of Primary Antibody Outcome
The second injection was administered to 92 to 96% of the adults and to 97 to 99% of the children (Table S2). The primary end point was assessed in more than 90% of the adults and children (Table S3).
Safety
Injection-Site Reactions and Symptoms in Adults
Injection-Site Reactions, Targeted Symptoms, and Serious Adverse Events.
Most injection-site reactions and symptoms were reported by participants in the 7 days after vaccination, were more frequent in the active-vaccine groups than in the placebo group, and were largely of grade 1 severity (Table 2 and Table S4). Through day 7, the incidence of injection-site reaction was 9% in the Ad26–MVA group, 22% in the pooled rVSV groups, and 5% in the placebo group. The percentages of participants who reported symptoms through day 7 were 50%, 66%, and 44%, respectively. By day 14, the frequency of injection-site reactions and symptoms were similar in the Ad26–MVA group, the pooled rVSV groups, and the placebo group (Table 2). Similar trends were observed after the second or booster vaccinations (Table 2 and Table S5). By month 3, the percentage of participants reporting injection-site reactions and symptoms was similar in the vaccine groups and the placebo group (Table 2).
Serious Adverse Events and Deaths through 12 Months in Adults
Serious adverse events were reported in 14 of 396 adults (4%) in the Ad26–MVA group, 6 of 395 (2%) in the rVSV group, 1 of 197 (1%) in the rVSV–booster group, and 5 of 412 (1%) in the pooled placebo group (Table 2 and Table S6). Overall, 6 adults died: 2 in the Ad26–MVA group (from septic shock and myocardial infarction, in 1 participant each), 3 in the rVSV group (from sepsis, HIV infection, and appendicitis, in 1 each), and 1 in the placebo group (from malaria). None of the deaths were judged by the site investigator to be related to the vaccine or placebo.
Injection-Site Reactions and Symptoms in Children
As in adults, most injection-site reactions and symptoms that were reported in children occurred in the week after the first injection (through day 7), were more frequent in the vaccine groups than in the placebo group, and were largely of grade 1 severity (Table 2 and Table S7). Through day 7, the incidence of injection-site reaction was 15% in the Ad26–MVA group, 21% in the pooled rVSV groups, and 5% in the placebo group. Symptoms through day 7 were reported in 48%, 60%, and 43% of the participants, respectively. Through day 14, the incidences of injection-site reactions and symptoms were similar in the Ad26–MVA group, the pooled rVSV groups, and the placebo group. Similar trends were observed after the second injection (Table 2 and Table S8). By month 3, the incidence of symptoms was similar in the vaccine groups and the placebo group. Differences between the vaccine groups and the placebo group were similar in each age group (Tables S9 through S13).
Serious Adverse Events and Deaths through 12 Months in Children
Serious adverse events were reported in 5 of 403 children (1%) in the Ad26–MVA group, in 9 of 407 (2%) in the rVSV group, in 3 of 202 (1%) in the rVSV–booster group, and in 8 of 389 (2%) in the placebo group (Table 2 and Table S14). Overall, 5 children died: 3 children in the rVSV group (from sudden unexplained death, drowning, and fever of unknown origin, in 1 participant each) and 2 in the placebo group (from cardiopulmonary failure after malaria and from suspected meningitis, in 1 participant each). None of the deaths were judged by the site investigator to be related to the vaccine or placebo.
Antibody Responses
Geometric Mean Concentrations and Antibody Response According to Follow-up Visit. 
Antibody Response in Adults and Children (Geometric Mean Concentrations), According to Trial Visit.
The geometric mean concentration was based on a log10 concentration with baseline log10 titer and trial site as covariates. Antibody response was defined as an antibody concentration of at least 200 enzyme-linked immunosorbent assay units (EU) per milliliter and an increase from baseline in the antibody concentration by at least a factor of 4. Geometric mean concentrations are shown. 𝙸 bars indicate 95% confidence intervals.
Among adults at month 12, the percentages of participants with an antibody response and the geometric mean concentrations were 41% and 401 EU per milliliter in the Ad26–MVA group, 76% and 992 EU per milliliter in the rVSV group, 81% and 1037 EU per milliliter in the rVSV–booster group, and 3% and 93 EU per milliliter in the placebo group (P<0.001 for all comparisons of vaccine with placebo) (Table 3). In both the Ad26–MVA and rVSV–booster groups, the percentages of participants with an antibody response and the geometric mean concentrations were greatest 7 days after receipt of the second or booster vaccination (day 63) and at month 3. By month 12, the percentage of participants with a response and the geometric mean concentrations had decreased to the levels that had been observed before the receipt of the second or booster vaccination (Table 3 and Figure 1A).
Among children at month 12, the percentages of participants with a response and the geometric mean concentrations were 78% and 828 EU per milliliter in the Ad26–MVA group, 87% and 1415 EU per milliliter in the rVSV group, 93% and 1745 EU per milliliter in the rVSV–booster group, and 4% and 67 EU per milliliter in the placebo group (P<0.001 for all comparisons of vaccine with placebo) (Table 3 and Tables S17 and S18). In each of the vaccine groups, the percentages of participants with a response and the geometric mean concentrations at month 12 were greater among children than among adults. Among children, as was observed among adults, both the Ad26–MVA and rVSV–booster groups had their greatest percentage of participants with a response and highest geometric mean concentrations at visits shortly after the second or booster vaccination (day 63) and at month 3. By month 12, the percentage of participants with a response and the geometric mean concentrations had decreased to the levels observed immediately before the receipt of the second or booster vaccination (Table 3 and Figure 1B).
Distribution of Increase in Antibody Concentration from Baseline.
These results were similar in an analysis that excluded participants with a baseline (prevaccination) antibody concentration of at least 200 EU per milliliter, since these persons may have had an asymptomatic infection in a previous outbreak (Table S19). Results were also similar to those in the primary analysis when we imputed the assay developer’s lower limit of quantification (66.96 EU per milliliter) for baseline and follow-up antibody concentrations below that level (Table S20). To facilitate consideration in the context of previous studies, we determined antibody responses for a range of differences in the factor increase (Table 4) and follow-up antibody levels above a specific level (Tables S21 and S22).
Subgroup analyses according to laboratory and country are provided in the Figures S3 through S8 and Tables S23 and S24. Although there were differences in the geometric mean concentrations among the countries and laboratories, all the differences in response between the vaccine group and the placebo group were large and followed a similar pattern of geometric mean concentrations over follow-up among both the adults and children, regardless of country or laboratory.
