Trial population
Between September 20 and November 10, 2022, 561 participants were screened, of which 20 were ineligible to take part in the study and one withdrew consent before vaccination. 270 participants received Ad5-nCoV-IH, of which 16 dropped out early and 254 completed the trial among whom the mean age was 34.5 (standard deviation (SD), 10.73) years, with 61.85% male. 270 participants randomized to BNT-IM, with one withdrawn consent before administration of vaccines. 269 participants received BNT-IM, of which 20 dropped out early and 249 completed the trial among whom a mean age was 34.4 (SD, 10.60) years, with 59.48% male (Fig. 1; Table 1). The distribution of gender, age, and race among these two groups was well-balanced. Participants of 25.56% receiving AZ + AZ + X (AZ: COVID-19 Vaccine, ChAdOx1-S, Recombinant; X: the first booster dose of any brand), 39.26% receiving mRNA+mRNA+X (mRNA: tozinameran vaccine), 32.58% receiving ICV + ICV + X (ICV: COVID-19 Vaccine (Vero Cell), Inactivated), and 2.59% having received a COVID-19 vaccine outside these categories in the Ad5-nCoV-IH group. In comparison, the BNT-IM group had percentages of 18.59%, 42.01%, 36.43%, and 2.97% for similar vaccine types, respectively. The median duration from the primary series to the first booster for all participants was 168 days, and that from the first to the second booster was 274 days (Table 1).
Fig. 1: Participant disposition flow chart.
BNT-IM Intramuscular tozinameran vaccine, Ad5-nCoV-IH Recombinant Covid-19 Vaccine (Adenovirus Type 5 Vector) for Inhalation.
Table 1 Baseline characteristics of analysis population
Primary outcomes
All 539 participants underwent primary immunogenicity assessments, and persistent immunogenicity assessments (at additional visits of week 12 and 24) were conducted on 139 participants in the experimental group and 141 in the control group. Results based on the per-protocol set (PPS) in Fig. 2a (refer to Supplementary Table 1 and 2 for PPS and mITT set, respectively) revealed that the GMC of S-RBD IgG antibodies against the Wild-type in the Ad5-nCoV-IH group rose from 1273.454 (95% CI, 1111.45-1459.07) at the start to 1555.533 (95% CI, 1377.17-1757.00) by day 14, representing a geometric mean increase (GMI) of 1.23 (SD, 2.09). In comparison, the BNT-IM group rose from 1231.701 (95% CI, 1062.93-1427.27) before vaccination to 6975.264 (95% CI, 6407.51-7593.33), with the GMI was 5.74 (SD, 2.74) (Supplementary Table 3). The seroconversion rate (SCR) were 17.10% (95% CI, 12.80%- 22.14%) for the Ad5-nCoV-IH group and 89.02% (95% CI, 84.61%-92.52%) for the BNT-IM group at 14 days (Supplementary Table 4). Analysis showed that the GMC ratio between the Ad5-nCoV-IH and BNT-IM groups was 0.22 (97.5% CI, 0.19-0.26) and a difference in SCR was -71.91% (95% CI, -78.63%-65.19%), suggesting that the Ad5-nCoV-IH vaccine induced lower humoral immune responses compared to the BNT-IM vaccine at 14 days post-vaccination (Supplementary Table 5). However, unlike the rapid response seen with BNT-IM at day 14 post-vaccination, the Ad5-nCoV-IH group reached its peak antibody levels at day 28, with a proportion of 27.61% (95% CI, 22.35%-33.38%). By day 168 post-vaccination, both vaccine groups showed similar levels of antibodies with no significant difference.
Fig. 2: Immune responses elicited by a second booster vaccination.
Ad5-nCoV-IH Recombinant Covid-19 Vaccine (Adenovirus Type 5 Vector) for Inhalation, BNT-IM intramuscular tozinameran vaccine, Anti-spike RBD IgG anti-spike receptor binding domain immunoglobulin G, GMC geometric mean concentration, GMT geometric mean titer, PNA pseudo neutralization assay, SIgA secretory Immunoglobulin A. a GMC of S-RBD IgG antibodies (wild-type). b GMT of Pseudovirus Neutralizing Antibodies (BA.4/5). The blue bar is Ad5-nCoV-IH group among which primary immunogenicity assessments were performed at Day 0, Day 14, and Day 28 for 270 participants, and persistent immunogenicity assessments were performed at Day 84 and Day 168 for 139 participants. The orange bar is BNT-IM group among which primary immunogenicity assessments were performed at Day 0, Day 14, and Day 28 for 268 participants, and persistent immunogenicity assessments were performed at Day 84 and Day 168 for 141 participants. c GMC of SIgA antibodies against Spike protein from BA.2.75, BA.2.75.2, and BA.4.6. The blue bar is Ad5-nCoV-IH group among which immunogenicity assessments were performed at Day 0, Day 14, Day 28, and Day 84 for 44 participants. The orange bar is BNT-IM group of which 44 participants’ immunogenicity assessments were performed at Day 0, Day 14, Day 28 and Day 84. * Indicates statistically significant differences between the Ad5-nCoV-IH group and the BNT-IM group (p < 0.05).
According to the PPS-based analysis, the GMC ratio of the anti-spike RBD IgG antibodies (wild-type) between the Ad5-nCoV-IH and BNT-IM groups was 0.22 (95% CI: 0.19, 0.26) at 14 days after vaccination, with the lower limit of the 97.5% CI < 0.67, which concluded that the antibody GMC non-inferiority hypothesis is not valid.
Secondary outcomes
Neutralizing antibodies (NAb) against Omicron BA.4/5 pseudovirus exhibited a pattern similar to that of S-RBD IgG antibodies. Participants in the experimental group showed a modest increase in geometric mean titre (GMT) from 286.92 (95% CI, 244.82–336.26) before vaccination to 395.3 (95% CI, 340.64–458.62) at day 14, peaking at 574.9 (95% CI, 457.54–722.31) on day 84, while the control group exhibited a sharp rise from 319.38 (95% CI, 270.75–376.75) pre-vaccination to a peak level of 1338.8 (95% CI, 1200.13–1493.50) by day 14, decreasing dramatically to 752.8 (95% CI, 616.45–919.24) at day 84 (Fig. 2b). Supplementary Table 6 and 7 showed the data for mITT and PPS set. The GMIs and SCRs were shown in Supplementary Table 8 and Supplementary Table 9. Collectively, the differences in GMT, GMI and SCR between the two groups were statistically significant at 14 and 28 days but not at 84 and 168 days after vaccination.
Considering the potential of respiratory tract delivery to induce mucosal immunity, our investigation extended to examine SIgA antibodies targeting the spike protein of the wild-type and also nine Variants of Concern (VOCs), B.1.1.529, BA.2.75, BA.2.75.2, BA.4.6, BA.5, BF.7, BQ.1, BQ.1.1, and XBB.1, involving a total of 88 participants with consecutive participant IDs (excluding those with positive COVID-19 cases post-vaccination). Seven participants were excluded due to invalid results, resulting in 42 from the Ad5-nCoV-IH group and 39 from the BNT-IM group. As depicted in Fig. 2c, besides Ad5-nCoV-IH demonstrated certain advantage and significant difference compared to BNT-IM in GMC of SIgA against BA.2.75 (32.072 (95% CI, 21.63–47.56) vs. 19.217 (95% CI, 14.49–25.49)), BA2.75.2 (37.085 (95% CI, 25.06–54.88) vs. 22.696 (95% CI, 16.99–30.33)) and BA.4.6 (42.832 (95% CI, 29.02–63.23) vs. 26.371 (95% CI, 19.70–35.30)) respectively on Day 28 post-vaccination (Supplementary Table 10), a more extensive advantage with significant difference was observed as well in the GMI encompassing BA.2.75, BA.2.75.2, BA.4.6, BA.5, BF.7, BQ.1, BQ.1.1 and XBB.1, respectively on Day 28 (Supplementary Table 11).
In the Ad5-nCoV-IH group, 127 participants (47.04%) experienced 817 episodes of adverse event (AE) with 106 participants (39.26%) having 515 vaccine related AE, referred to as adverse drug reactions (ADRs), within 28 days post-vaccination. In the BNT-IM group, 189 participants (70.26%) experienced 1171 episodes of AEs with 174 participants (64.68%) reporting 892 ADRs. Although statistical tests for safety parameters were not predefined in the protocol, we conducted them and found statistically significant differences in the number of AEs and ADRs between the two groups (P < 0.001), with the BNT-IM group showing a higher incidence (Fig. 3 for ADRs and Supplementary Table 12 for AEs). As Ad5-nCoV-IH was administered through oral inhalation, oropharyngeal pain, pharyngeal swelling, dysphonia, dry mouth, and stomatitis were specifically attributed to local ADR in the Ad5-nCoV-IH group. In terms of local ADRs, oropharyngeal pain was the most frequently reported in the Ad5-nCoV-IH group at 12.96%. In terms of other local ADRs in this group, dysphonia (9.26%), dry mouth (8.89%), pharyngeal swelling (5.56%), and stomatitis (3.70%) were frequently observed, with only dysphonia and pharyngeal swelling showing a statistically significant difference (P < 0.05) when compared to the BNT-IM group, despite being categorized as systemic reactions. For the BNT-IM group, vaccination site pain was the predominant local ADR at 57.62%. The occurrence rates of pruritus, swelling, erythema, induration, and rash as local ADRs were 5.20%, 4.83%, 1.49%, 1.49%, and 0.37%, respectively. There was a case from the Ad5-nCoV-IH group where vaccination site pain was reported as a local reaction, and the incidence was 0.37%. Ad5-nCoV-IH resulted in fewer systemic ADRs, showing lower frequencies for headache (20.74%), fatigue (18.15%), myalgia (12.22%), insomnia (4.07%), pain (1.48%), and pruritus (1.48%). In comparison, the BNT-IM group reported higher numbers for these systemic ADRs, with rates of 30.86%, 29.00%, 31.23%, 10.04%, 8.92%, and 4.46%, respectively. However, there were no significant differences in the incidence of grade 3 or higher ADRs (P = 0.788) and unsolicited ADRs (P = 0.765). No vaccine related serious adverse events (SAEs), adverse event of special interest (AESIs), withdrawal-related events, or deaths occurred (Supplementary Table 13 and Supplementary Table 14).
Fig. 3: Solicited and unsolicited ADRs emerged in the 0 to 28 Days post booster vaccination (SS).
Percentage of participants experienced ADR(s). The analysis was based on the safety cohort (SS), included all randomized participants who received the booster vaccination and had at least one evaluation data available (270 participants from the Ad5-nCoV-IH group and 269 participants from the BNT-IM group). *P < 0.05, **P < 0.001.
Starting from 14 days after the second booster vaccination, there were 22 confirmed COVID-19 cases in the Ad5-nCoV-IH group compared to 26 cases in the BNT-IM group, with a median follow-up period of 60 days post-vaccination, resulting in an adjusted vaccine efficacy (aVE of 0.19 (95% CI, -0.43–0.55). All positive cases in this study reported at least one symptom, such as sore throat, cough, or chills, according to participant questionnaires for COVID-19. There was no significant difference for efficacy of the Ad5-nCoV-IH and IMB-BNT at follow-up periods of 90 days with aVE of 0.20 (95% CI, -0.35-0.53) and 120 days with aVE of 0.22 (95% CI, -0.30-0.53), as the study was underpowered for this endpoint. The number of COVID-19 cases became comparable between the two groups by the end of the study, yielding an aVE of 0.07 (95% CI, -0.49-0.42) with a median follow-up period of 169 days (Table 2 and Supplementary Fig. 1). An exploratory analysis of vaccine effectiveness against breakthrough infections was assessed based on participants’ infection and vaccination histories (Supplementary Table 15). The study was not designed to detect a statistically significant difference for this endpoint. Those with a history of infection (COVID-19 confirmed by PCR/ rapid test kit (RTK) test) in the Ad5-nCoV-IH group consistently demonstrated no statistical difference with aVE values of 0.31 (95% CI, -0.57-0.7), 0.32 (95% CI, -0.42-0.67), 0.34 (95% CI, -0.34-0.67), and 0.12 (95% CI, -0.64-0.53) compared to the BNT-IM group at median follow-up periods of 60, 90, 120 and 169 days, respectively. Individuals without a prior infection history exhibited similar efficacy between the two groups. Another exploratory analysis of aVE was performed based on vaccination history, when Ad5-nCoV-IH served as a second booster dose following previous vaccination of mRNA+mRNA+X, it demonstrated no statistical difference in aVE compared to BNT-IM over a median follow-up period of 60 days (0.38, 95% CI, -0.56-0.75), 90 days (0.33, 95% CI, -0.53-0.71) and 120 days (0.33, 95% CI, -0.53, 0.71). Among individuals with a vaccination history of ICV + ICV + X, Ad5-nCoV-IH exhibited an aVE of 0.04 (95% CI, -1.88-0.68) at a 60-day follow-up, 0.16 (95% CI, -1.3-0.69) at 90-day follow-up, and 0.29 (95% CI, -0.78- 0.71) at 169-day follow-up. Overall, the efficacy results indicated that the Ad5-nCoV-IH group showed similar protective efficacy in preventing breakthrough infections.
Table 2 Vaccine efficacy of recombinant COVID-19 vaccine (Ad5-nCoV-IH) in preventing virologically confirmed COVID-19 cases at 14 days after vaccination
We conducted supplementary examinations on the genetic composition of circulating variants throughout the entire study duration. Out of the nasopharyngeal samples from 69 endpoint cases, 64 were eligible for sequencing. All identified variants were of the Omicron lineage, including BA.2.10.1, BA.2, BA.5.2, XBB.1, BM.1.1, BN.1.1, BQ.1.1, and others. Additionally, eight cases remained unassigned (Supplementary Table 16).
