Heterologous AD5-nCOV plus CoronaVac versus homologous CoronaVac vaccination: a randomized phase 4 trial

Abstract

The emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants and the waning of vaccine-elicited neutralizing antibodies suggests that additional coronavirus disease 2019 (COVID-19) vaccine doses may be needed for individuals who initially received CoronaVac. We evaluated the safety and immunogenicity of the recombinant adenovirus type 5 (AD5)-vectored COVID-19 vaccine Convidecia as a heterologous booster versus those of CoronaVac as homologous booster in adults previously vaccinated with CoronaVac in an ongoing, randomized, observer-blinded, parallel-controlled phase 4 trial ( NCT04892459 ). Adults who had received two doses of CoronaVac in the past 3-6 months were vaccinated with Convidecia (n = 96) or CoronaVac (n = 102). Adults who had received one dose of CoronaVac in the past 1-3 months were also vaccinated with Convidecia (n = 51) or CoronaVac (n = 50). The co-primary endpoints were the occurrence of adverse reactions within 28 d after vaccination and geometric mean titers (GMTs) of neutralizing antibodies against live wild-type SARS-CoV-2 virus at 14 d after booster vaccination. Adverse reactions after vaccination were significantly more frequent in Convidecia recipients but were generally mild to moderate in all treatment groups. Heterologous boosting with Convidecia elicited significantly increased GMTs of neutralizing antibody against SARS-CoV-2 than homologous boosting with CoronaVac in participants who had previously received one or two doses of CoronaVac. These data suggest that heterologous boosting with Convidecia following initial vaccination with CoronaVac is safe and more immunogenic than homologous boosting.

Conflict of interest statement

X.W., J.W., J.L. and T.Z. are employees of CanSino Biologics. All other authors declare no competing interests.

© 2022. The Author(s).

Figures

Fig. 1. Consolidated Standards of Reporting Trials (CONSORT) flow diagram.

Six participants discontinued follow-up after vaccination. The reasons for dropout are withdrawn consents for participation. *Two participants were randomized to group A but were incorrectly administrated an inactivated vaccine and then classified into group B. On the other hand, another participant was only primed with one dose but was incorrectly classified into the population with two doses of prime vaccination and randomized to group A to receive one dose of AD5-based vaccine. We reclassified this participant into group C.

Fig. 2. Neutralizing antibodies to wild-type SARS-CoV-2 or the Delta variant before and after boosting.

GMTs of neutralizing antibodies to wild-type SARS-CoV-2 (a) or the Delta variant B1.617.2 (b). GMFI of neutralizing antibodies to wild-type SARS-CoV-2 (c) or the Delta variant B1.617.2 (d). Seroconversion of neutralizing antibodies to wild-type SARS-CoV-2 (e) or the Delta variant B1.617.2 (f). Error bars indicate 95% CIs. n, the number of participants included the intervention modified intention-to-treat cohort; seroconversion, proportion of participants with at least a fourfold increase in post-vaccination antibody levels compared to levels before the booster vaccination. Group A, primed with two doses of CoronaVac and given one dose of Convidecia (n = 96); group B, primed with two doses of CoronaVac and given one dose of CoronaVac (n = 102); group C, primed with one dose of CoronaVac and given one dose of Convidecia (n = 51); group D, primed with one dose of CoronaVac and given one dose of CoronaVac (n = 50). The analysis was based on the intervention modified intention-to-treat cohort. Measurements on day 0 were taken immediately before vaccination. The WHO reference (1,000 IU ml−1 in serum) is equivalent to a live viral neutralizing antibody titer of 1:320 against wild-type SARS-CoV-2 and 1:80 against the Delta variant B.1.617.2. P values result from comparison between the two treatment groups using t-tests for log-transformed antibody titers or two-sided χ2 tests for categorical data (group A versus group B and group C versus group D). No adjustments were made for multiple comparisons (group A versus group B and group C versus group D). For (e,f), the statistics are proportions of participants with seroconversion after the vaccination. **P < 0.005, ****P < 0.0001.

Fig. 3. RBD- or N-specific antibody levels before and after boosting.

GMTs of anti-RBD IgG (a) and anti-N IgG (b). GMFI in anti-RBD IgG (c) and anti-N IgG (d) levels. Seroconversion of anti-RBD IgG (e) and anti-N IgG (f). Data are GMT (95% CI), GMFI (95% CI) or seroconversion (95% CI). Error bars indicate 95% CIs. n, the number of participants included the intervention modified intention-to-treat cohort; seroconversion, proportion of participants with at least a fourfold increase in post-vaccination antibody levels compared to levels before the booster vaccination. Measurements on day 0 were taken immediately before vaccination. Group A, primed with two doses of CoronaVac and given one dose of Convidecia (n = 96); group B, primed with two doses of CoronaVac and given one dose of CoronaVac (n = 102); group C, primed with one dose of CoronaVac and given one dose of Convidecia (n = 51); group D, primed with one dose of CoronaVac and given one dose of CoronaVac (n = 50). The WHO reference (1,000 binding antibody units (BAU) ml−1 in serum) is equivalent to an RBD-specific IgG ELISA antibody titer of 1:2,430. P values result from comparison between the two treatment groups using t-tests for log-transformed antibody titers or two-sided χ2 tests for categorical data (group A versus group B and group C versus group D). No adjustments were made for multiple comparisons. For (e,f), the statistics are proportions of participants with seroconversion after the vaccination. ****P < 0.0001.

Fig. 4. SARS-CoV-2 spike-specific T cell cytokine responses before and after boosting.

TH1/TH2 ratios (f) were calculated by summing IFN-γ (a) and TNF-α (b) cytokine levels and then dividing by the sum of IL-4 (c), IL-5 (d) and IL-13 (e) cytokine levels. Data are the median (quartile 1, quartile 3) of positive spot counts per 106 PBMCs. Horizontal bars show the median, and error bars indicate IQR. Group A, primed with two doses of CoronaVac and given one dose of Convidecia (n = 24); group B, primed with two doses of CoronaVac and given one dose of CoronaVac (n = 25); group C, primed with one dose of CoronaVac and given one dose of Convidecia (n = 16); group D, primed with one dose of CoronaVac and given one dose of CoronaVac (n = 15). Cytokine-secreting T cells were background corrected for unstimulated cells, and values lower than 0 were considered negative. Samples of PBMCs were collected from the first 50 and 30 participants in the three-dose and two-dose regimen cohorts, respectively, and included in the analysis. Apparent discrepancies between the numbers of data points presented in the figures and the numbers of participants in the groups are due to overlapping dots. P values result from comparison between the two treatment groups using Wilcoxon rank-sum tests (group A versus group B and group C versus group D). No adjustments were made for multiple comparisons. *P < 0.05, **P < 0.005.

Extended Data Fig. 1. GMT of neutralizing antibodies to wild-type SARS-CoV-2 at day 14 after vaccination stratified by age-subgroups and sex groups.

(a) GMTs of neutralizing antibodies to wild-type SARS-CoV-2 stratified by age-subgroups in Group A and Group B. (b) GMTs of neutralizing antibodies to wild-type SARS-CoV-2 stratified by age-subgroups in Group C and Group D. (c) GMTs of neutralizing antibodies to wild-type SARS-CoV-2 stratified by sex-subgroups in Group A and Group B. (d) GMTs of neutralizing antibodies to wild-type SARS-CoV-2 stratified by sex-subgroups in Group C and Group D. Data are GMT (95% CI). Error bars indicate 95% CIs. Group A: primed with two doses of CoronaVac + Convidecia (n = 96); Group B: primed with two doses of CoronaVac + CoronaVac (n = 102); Group C: primed with one dose of CoronaVac + Convidecia (n = 51); Group D: primed with one dose of CoronaVac + CoronaVac (n = 50). The p values are the results of comparison using T test for log-transferred antibody titers by age-subgroups and sex groups in each treatment group. Measurements were on day 0 were taken immediately before vaccination. GMT = geometric mean titer. GMFI = geometric mean fold increase. ***P value

Extended Data Fig. 2. Receptor binding domain (RBD)-binding IgG isotypes before and after receiving a heterogeneous or homologous boost vaccine.

(a) GMTs of anti-RBD IgG1 isotypes, (b) IgG2 isotypes, (c) IgG3 isotypes and (d) IgG4 isotypes. Data are GMT (95% CI). Error bars indicate 95% CIs. Group A: primed with two doses of CoronaVac + Convidecia (n = 96); Group B: primed with two doses of CoronaVac + CoronaVac (n = 102); Group C: primed with one dose of CoronaVac + Convidecia (n = 51); Group D: primed with one dose of CoronaVac + CoronaVac (n = 50). The p values are the results of comparison between the two treatment groups using T test for log-transferred antibody titers (Group A vs. Group B, and Group C vs. Group D). All the paired data of RBD-binding antibodies from participants are included in the analysis. The discrepancies between the numbers of data points presented in the figures and the numbers of participants in the groups are due to the overlapping of the dots. **P value

Extended Data Fig. 3. Correlations between live SARS-CoV-2 neutralizing antibodies against the wild-type strain and receptor binding domain (RBD) antibodies by vaccine regimens at day 14 and 28 post-vaccination.

(a) Correlations between live SARS-CoV-2 neutralizing antibodies against the wild-type strain and receptor binding domain (RBD) antibodies by vaccine regimens at day 14 and day 28 (b) in Group A and Group B. (c) Correlations between live SARS-CoV-2 neutralizing antibodies against the wild-type strain and receptor binding domain (RBD) antibodies by vaccine regimens at day 14 and day 28 (d) in Group C and Group D. Data are GMT (95% CI), GMFI (95% CI), and seroconversion (95% CI). Group A: primed with two doses of CoronaVac + Convidecia (n = 96); Group B: primed with two doses of CoronaVac + CoronaVac (n = 102); Group C: primed with one dose of CoronaVac + Convidecia (n = 51); Group D: primed with one dose of CoronaVac + CoronaVac (n = 50). Pearson correlation coefficients (95% CIs) are presented for each vaccine schedule. All the paired data of neutralizing antibodies and RBD-binding antibodies from participants are included in the analysis. The discrepancies between the numbers of data points presented in the figures and the numbers of participants in the groups are due to the overlapping of the dots.