Safety and immunogenicity of fractional COVID-19 vaccine doses in Nigerian adults: A randomized non-inferiority trial

Abstract

Fractional dosing of vaccines is a viable strategy to extend COVID-19 vaccine supplies in resource-constrained settings. We did a triple-blinded, multi-site, randomized non-inferiority trial in Nigeria (PACTR202206754734018). Adults 18-65 years received full, half, or quarter primary doses of ChAdOx1 or Ad26.COV2.S, or full vs half doses of BNT162b2. Primary study outcome was seroconversion defined as ≥ 2.5-fold rise in anti-Spike IgG geometric-mean fold rise (GMFR) at day 28. A total of 1894 participants were enrolled between June 21, 2022, and January 25, 2023. 320 participants in the fractional dose group and 220 in the standard dose group completed follow-up and were included in the analysis. Seropositivity at baseline was high, at 68% (365/539). Seroconversion was comparable between standard and fractional doses (p = 0.822). For ChAdOx1, 31% achieved seroconversion at standard dose (16/52), 28% at half-dose (15/53), and 34% in quarter-dose (18/53). For Ad26.COV2.S, the proportions were 27% (28/105), 32% (22/68), and 30% (21/71) respectively. For BNT162b2, the proportions were 43% (27/63) and 39% (29/75) for standard- and half-dose. Serum neutralization showed ≥ twofold response across dosing. There were no serious adverse events. Fractional vaccine doses generated non-inferior immune responses compared to standard doses in the context of previous COVID-19.Protocol Registration: The protocol was registered with the Pan African Clinical Trials Registry (PACTR) PACTR202206754734018.

Fig. 1

Study flow chart and disposition of patient recruitment . a) CONSORT diagram of the enrolment and screening, randomisation, and follow-up of participants randomized to receive full or fractionated doses of ChadOx1, Ad26.COV2.S and BNT162B2 vaccines. b) Disposition of study population showing timepoints of vaccination and study visits. * Participants randomized to receive full doses of AD26.COV2.S only received one dose as per standard recommendation and those randomized to half and quarter doses received booster doses at 12 weeks.

Fig. 2

Primary outcome analysis and adverse events among the aggregated study participants (n = 540) a) proportion of participants across study arms with geometric mean fold rise (GMFR) ≥ 2.5 in serum anti-spike IgG titre at 28 days post-vaccination. Non-inferiority margin was set at 20% and point difference between the fractional and standard dose arms with 95% CIs using the Wilson score method. b) proportion of study participants reported at least one adverse event occurring within the 72 h following COVID-19 vaccine administration, irrespective of the study arm.

Fig. 3

Antibody responses to vaccination with BNT162b2 at full and half doses. a): Plasma neutralization of pseudovirus against Wild type (Wu-1), BA.1, BA.2 and XBB after two doses of the BNT162b2 in Nigerian HIV-negative participants in Lagos Nigeria at three consecutive time points– baseline – F0 (before first-dose vaccination),F1 (1 month after 1st dose vaccination) and F3 (6 weeks post second dose at full doses (n = 8) and half doses (n = 8). Data points were compared using Wilcoxon test and shown as geometric mean titre (GMT) with 95% CI. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns = not significant. Fold change are represented above the horizontal comparative lines.

Fig. 4

Antibody responses to vaccination with ChadOX1 at full, half and quarter doses. a): Plasma neutralization of pseudovirus against Wild type (Wu-1), BA.1, BA.2 and XBB after two doses of the ChadOX1 in Nigerian HIV-negative participants in Lagos Nigeria at three consecutive timepoints– baseline – F0 (before first-dose vaccination), F1 (1 month after 1st dose vaccination) and F3 (6 weeks post second dose at full, (n = 8); half (n = 8) and quarter doses (n = 8). Data points were compared using Wilcoxon test and shown as geometric mean titre (GMT) with 95% CI. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns = not significant. Fold change are represented above the horizontal comparative lines.

Fig. 5

Antibody responses to vaccination with Ad26.COV2.S at full, half and quarter doses. a): Plasma neutralization of pseudovirus against Wild type (Wu-1), BA.1, BA.2 and XBB after two doses of the Ad26.COV2.S in Nigerian HIV-negative participants in Lagos, Nigeria at three consecutive time points– baseline – F0 (before first-dose vaccination), F1 (1 month after 1st dose vaccination) and F3 (6-weeks post second dose at full, (n = 6); half (n = 8) and quarter doses (n = 8). Data points were compared using Wilcoxon test and shown as geometric mean titre (GMT) with 95% CI. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001; ns = not significant. Fold change are represented above the horizontal comparative lines.

Fig. 6

a). Fold-change following two doses of fractional doses of Ad26.COV2.S, ChadOX1 and BNT162b2. Plasma neutralization of pseudovirus against Wild type (Wu-1) fold-change following two doses of the Ad26.COV2.S, ChadOX1 and BNT162b2 doses either as full or quarter or half fractionated doses. Fold Change is shown above vaccine doses. b). Antibody responses to vaccination with BNT162b2, Ad26.COV2.S, ChadOX1 at full, half and quarter doses. Radar plot showing as geometric mean titre (GMT) plasma neutralization of pseudovirus against Wild type (Wu-1), BA.1, BA.2 and XBB after two doses. GMT data is shown in log scale.