One vs. two COVID-19 vaccine doses in previously infected children: antibody responses persisted, but noninferiority was not demonstrated
Background: Comparing dose schedules in previously infected children
A clinical trial comparing one dose of BNT162b2 with two doses in children aged 5–11 years who had previously had COVID-19 did not establish that one dose was noninferior. Both groups showed sustained antibody responses, but the trial ended early and lacked the statistical power needed for the comparison.
The CoVacc investigators sought to determine whether one dose was noninferior to two doses for immunogenicity in previously infected children, while offering a better safety and reactogenicity profile. Immunogenicity means the extent to which a vaccine elicits an immune response. The primary endpoint was neutralizing antibody levels, not actual infections or hospitalizations. The results were published in Vaccine.
This was an open-label, randomized phase 2 trial in which participants and investigators knew the assigned schedule. Participants were allocated 1:1, and the interval between doses in the two-dose group was 3–12 weeks. From May 2022 through January 2024, 31 children were enrolled in three European countries. The median age was 9 years (interquartile range, 7–10 years); 16 received one dose and 15 received two. The trial ended early as interest in COVID-19 vaccination among eligible children declined.
Key findings: Similar antibody averages, noninferiority not established
The primary endpoint was the geometric mean ratio (GMR) comparing geometric mean titers (GMTs) of neutralizing antibodies against wild-type SARS-CoV-2 28 days after vaccination. The noninferiority margin was prespecified as a 1.5-fold difference in GMT. Similar averages alone cannot establish noninferiority; the uncertainty around the estimate must also fall within the defined acceptable range.
At day 28, the GMT was 1,715.5 IU/mL in the one-dose group (95% confidence interval, 1,064.2–2,765.4) and 1,801.1 IU/mL in the two-dose group (95% confidence interval, 1,357.9–2,388.9). The GMR was 0.9 (95% confidence interval, 0.5–1.6), so noninferiority of one dose was not demonstrated. This does not establish that one dose was inferior or confirm that the schedules were equivalent.
In both groups, antibody levels remained above 100 IU/mL at 6 and 12 months. Separately, at day 28, the study reported high titers of immunoglobulin G (IgG) antibodies binding to the wild-type receptor-binding domain (RBD), as well as neutralizing antibody responses against the BA.5 variant in both groups. The abstract alone does not establish whether BA.5 neutralization persisted to 12 months. Nor was 100 IU/mL presented as a threshold that guarantees clinical protection.
Meaning and outlook: A small trial points to booster potential
Within 7 days after vaccination, grade 2 or higher systemic adverse events were reported in 8 participants (53%) in the two-dose group and 5 (31%) in the one-dose group. Although the proportion was numerically lower with one dose, the between-group comparison had a P value of 0.18. This difference alone cannot support a conclusion that one dose is safer. The investigators considered the vaccine well tolerated in both groups.
The investigators interpreted the magnitude and persistence of the antibody responses as consistent with the possibility that one dose could act as a humoral immune booster in previously infected children. The results do not directly demonstrate a mechanism involving memory B-cell reactivation. With only 31 participants and early trial termination, the evidence is insufficient to determine whether clinical protection is maintained when the number of doses is reduced. The available abstract does not provide detailed analyses or the number of participants who completed follow-up, which also calls for caution when interpreting long-term results.
The study design and results are based on the provided CoVacc paper citation and abstract.
BACKGROUND: This open-label randomised phase 2 study aimed to determine whether a single versus two-dose BNT162b2 primary vaccination regimen in children 5 to 11 years old with prior SARS-CoV-2 infection was non-inferior in terms of immunogenicity and superior in terms of safety and reactogenicity. METHODS: Participants were randomly assigned (1:1) to receive either one or two doses, spaced 3 to 12 weeks. The primary endpoint was geometric mean ratio (GMR) of neutralizing antibodies against wild-type SARS-CoV-2 at 28 days post-vaccination with non-inferiority margin defined as a 1.5-fold change in geometric mean titers (GMT). Secondary endpoints included safety and reactogenicity profile and immunogenicity up to 12 months against wild-type and Variants of Concern (VOCs). RESULTS: In total 31 participants from 3 European countries (median age 9, IQR 7-10) were enrolled from May 2022 to January 2024, when the trial was prematurely terminated due to declining interest in COVID-19 vaccination among age-eligible children. Of these, 15 received two doses, and 16 received one. At day 28, GMT of neutralizing antibodies against wild-type SARS-CoV-2 was 1801.1 IU/mL (95% CI: 1357.9-2388.9) in the two-dose arm and 1715.5 IU/mL (95% CI: 1064.2-2765.4) in the single-dose arm. However, the non-inferiority of the single-dose could not be demonstrated (GMR: 0.9; 95% CI: 0.5-1.6). Titers remained above 100 IU/mL in both groups at 6 and 12 months. Both schedules elicited high anti-RBD IgG titers against wild-type and neutralizing titers against BA.5 variant at day 28. Eight participants (53%) in the two-dose arm and five (31%) in the single-dose reported a systemic adverse event grade ≥ 2 (P = 0.18) within 7 days of vaccination. CONCLUSIONS: Both regimens induced robust and sustained immune responses consistent with the possibility that, in children with prior infection, a single dose functions immunologically as a booster of the humoral response. However, the premature termination renders the
This study provides data for comparing antibody persistence and post-vaccination adverse events when evaluating COVID-19 vaccination schedules for children with a history of infection. Researchers planning follow-up comparative trials can use these antibody values and the uncertainty around the estimates to plan for adequate enrollment and follow-up to assess noninferiority. Recruitment challenges associated with declining interest in vaccination should also be considered in trial operations.
These results alone do not provide enough evidence for clinicians to tell families that one dose is sufficient or to change an individual child’s schedule. Observed persistence of antibody levels must be distinguished from evidence of protection against infection or hospitalization. The trial suggests that a single dose in previously infected children might act as a booster, while showing which uncertainties need to be resolved before considering fewer doses.
EMOTION_TAG: clinical