Safety and immunogenicity of heterologous boosting with orally administered aerosolized bivalent adenovirus type-5 vectored COVID-19 vaccine and B.1.1.529 variant adenovirus type-5 vectored COVID-19 vaccine in adults 18 years and older: a randomized, double blinded, parallel controlled trial

ABSTRACT

Vaccination strategies that can induce a broad spectrum immune response are important to enhance protection against SARS-CoV-2 variants. We conducted a randomized, double-blind and parallel controlled trial to evaluate the safety and immunogenicity of the bivalent (5×10¹⁰viral particles) and B.1.1.529 variant (5×10¹⁰viral particles) adenovirus type-5 (Ad5) vectored COVID-19 vaccines administrated via inhalation. 451 eligible subjects aged 18 years and older who had been vaccinated with three doses inactivated COVID-19 vaccines were randomly assigned to inhale one dose of either B.1.1.529 variant Ad5 vectored COVID-19 vaccine (Ad5-nCoVO-IH group, N=150), bivalent Ad5 vectored COVID-19 vaccine (Ad5-nCoV/O-IH group, N=151), or Ad5 vectored COVID-19 vaccine (5×10¹⁰viral particles; Ad5-nCoV-IH group, N=150). Adverse reactions reported by 37 (24.67%) participants in the Ad5-nCoVO-IH group, 28 (18.54%) in the Ad5-nCoV/O-IH group, and 26 (17.33%) in the Ad5-nCoV-IH group with mainly mild to moderate dry mouth, oropharyngeal pain, headache, myalgia, cough, fever and fatigue. No serious adverse events related to the vaccine were reported. Investigational vaccines were immunogenic, with significant difference in the GMTs of neutralizing antibodies against Omicron BA.1 between Ad5-nCoV/O-IH (43.70) and Ad5-nCoV-IH (29.25) at 28 days after vaccination (P=0.0238). The seroconversion rates of neutralizing antibodies against BA.1 in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were 56.00%, 59.60% and 48.67% with no significant difference among the groups. Overall, the investigational vaccines were demonstrated to be safe and well tolerated in adults, and was highly effective in inducing mucosal immunities in addition to humoral and cellular immune responses defending against SARS-CoV-2 variants.

Trial registration: Chictr.org identifier: ChiCTR2200063996.

KEYWORDS:

• Bivalent adenovirus type-5 vectored COVID-19 vaccine
• inhalation
• mucosal immunity
• SigA
• safety

Introduction

New SARS-CoV-2 variants continually emerge with ongoing COVID-19 pandemic around the world, the World Health Organization (WHO) has declared five notable variants of concerns (VOCs), including Omicron (B.1.1.529), which is highly transmissible and immune-evading [1]. Then, Omicron has given rise to multiple subvariants, with the first subvariant of Omicron dubbed BA.1 (B.1.1.529.1), detected in South Africa in November 2021. Since then, several other subvariants have been identified and studied, including BA.1.1, BA.2, BA.2.12.1, BA.2.7, XBB, BA.3, BA.4, BA.5, BF.7, BQ.1, and BQ.1.1. These subvariants exhibit distinct properties and have been associated with different epidemiological trends [2]. Although the BA.5 subvariant is globally dominant at this time, a diverse array of Omicron sublineages have arisen and are competing in the so-called “variant soup” [3]. For example, XBB and XBB.1 were first identified in India in mid-August and quickly became predominant in India, Singapore, and other regions in Asia. BQ.1 and BQ.1.1 evolved from BA.5, whereas XBB and XBB.1 resulted from recombination between two BA.2 lineages, BJ.1 and BA.2.75.

Despite numerous COVID-19 vaccines have been approved for emergency use, the rapid evolution of the variants escaped the protection offered by the primary generation of COVID-19 vaccines [4,5]. Existing COVID vaccines which developed on various technology platforms including inactivated vaccines, mRNA vaccines, and adenovirus vector-based vaccines exhibited good performance in protection against the wild-type SARS-CoV-2, however, antibodies induced by these vaccines showed significantly low neutralization effects on variants especially Omicron [6–9]. Given that vaccine remains the most effective weapon for preventing and controlling the spread of the virus, scientists focus on developing multivalent vaccines and seeking a better way to stimulate mucosal immunity to defense currently prevalent viruses as well as the new emerging variants. However, there are still too much uncertain immune responses to coronavirus infections to make it more durable and have a large breadth against variants [10,11].

Research has shown that vaccines given via intranasal route can elicit strong mucosal and systemic immune responses [12]. Recently, a novel vaccine delivery route by nebulizing the vaccine into mist followed by orally inhaling into rhesus macaques was established. Xu et al. have further confirmed the advantage of this orally administered method through the detection of exceptionally high anti-spike protein receptor binding domain (RBD) IgG levels, neutralizing antibodies for SARS-CoV-2 and Delta pseudovirus as well as secretory IgA (SIgA) antibody which regarded as a feature marker of mucosal immunity in bronchoalveolar lavage (BAL) [13].

Respiratory mucosal vaccines may be a potential strategy to induce human immunity combating the long-lasting pandemic as literature pointed out that this specific vaccination approach can stimulate all-around mucosal immunity against ancestor virus, current prevalent variants, and even new emerging VOCs [14]. CanSino BIO has conducted comprehensive clinical trials studying the safety and immunogenicity of aerosolized Ad5-nCoV vaccine, and results demonstrated that this innovative vaccination route could endow the vaccine with desirable safety, immunogenicity, and immunogenicity persistence profiles [15–17]. Among them, the results of our analysis of a phase 3 trial conducted at 15 centres in six provinces in China showed that a heterologous booster regimen of aerosolized Ad5-nCoV in healthy adults was safe and highly immunogenic, enhancing systemic and mucosal immunity against the Omicron subvariant [18]. And we also assess the safety and immunogenicity of heterologous booster immunization with orally administered Ad5 in children and adolescents, which showed that a safe and highly immunogenic profile against ancestral SARS-CoV-2 Wuhan-Hu-1 [19].

To combine the advantage of inhaled vaccination route with the extensive immunity of multivalent vaccine, Institute of Bioengineering of the Academy of Military Medicine of the Academy of Military Sciences and CanSino Biologics Co., Ltd. jointly developed inhaled monovalent B.1.1.529 variant Ad5 vectored COVID-19 vaccine, and bivalent Ad5 vectored COVID-19 vaccine by separately inserting S protein from wild-type and Omicron variant in adenovirus type-5 vectors and conducted a randomized and double-blind clinical trial, using them as the booster in healthy adults who have received 3 doses of inactivated vaccine compared with those boosting with Ad5 vectored COVID-19 vaccine. Here, we report the safety and immunogenicity including mucosal immune responses of the Ad5-nCoVO-IH and Ad5-nCoV/O-IH.

Materials and methods

Study design and participants

We did a randomized, double blinded, parallel controlled trial in accordance with the Declaration of Helsinki and Good Clinical Practice in Chongqing, China. This trial was registered in Chictr.org (ChiCTR2200063996). Adults aged 18 years and older who have received three doses of inactivated COVID-19 vaccine with an interval for ≥ 3 months between the last vaccination and those can provide negative PCR certificate within 7 days were eligible for screening. Volunteers with history of COVID-19 infection/illness and SARS, history of convulsion, epilepsy, encephalopathy, and severe neurological diseases in the past five years, known or suspected concomitant serious diseases judged by the investigator, including diabetes, thyroid disease, respiratory disease, tuberculosis, acute infection or active chronic disease, liver and kidney disease, cardiovascular disease, uncontrolled hypertension, malignant tumour, infection or allergic skin disease, and HIV infection were not able to join this study. Females of childbearing age with positive urine pregnancy test or lactating volunteers, and volunteers or their partners planning to become pregnant within 6 months were also excluded.

The protocol and informed consent were approved by the ethics committee of the Chongqing Center for Disease Control and Prevention (KY-2022-024-4). All participants gave written informed consent before screening.

Randomization and masking

The participants were randomly assigned (1:1:1) to receive one dose of B.1.1.529 variant Ad5 vectored COVID-19 vaccine (Ad5-nCoVO-IH), bivalent Ad5 vectored COVID-19 vaccine (Ad5-nCoV/O-IH), or Ad5 vectored COVID-19 vaccine (Ad5-nCoV-IH) with unique randomization number on each vial as the only identifiers. A blind code was generated according to a blocked randomization list (block size 3) using SAS statistical software. The randomization specialist imported the blind codes into the central randomization system (Interactive Web Response System IWRS) and use IWRS to manage the randomization. In the trial, subjects, investigators, project statisticians, and laboratory testers were masked to group allocation. Vaccine administrators and preparation staffs who were not involved in onsite work other than administering and preparing vaccines are non-blind. According to the random coding, the identified experimental vaccines were labelled and blinded. Personnel in charge of blinding printed copies of the three vaccine numbers pasted the corresponding label numbers on the corresponding vaccine boxes and vials, and finally placed them in order to complete the blinding work.

Procedures

Vaccines were stored and transported at 2–8°C away from light, and were supplied at a concentration of 5 × 10¹⁰ viral particles per mL as a liquid formulation with a volume of 0.5 mL. The Ad5 vectored COVID-19 vaccine contained replication-defective Ad5 vectors expressing the full-length spike gene of wild-type SARS-CoV-2, Wuhan-Hu-1 (GenBank accession number YP_009724390). B.1.1.529 variant Ad5 vectored COVID-19 vaccine contained replication-defective Ad5 vectors expressing spike gene of B.1.1.529 with the sequence induced of furrin cleavage site mutation and two proline (2P) mutations. (GISAID ID: EPI_ISL_6640917). Bivalent Ad5 vectored COVID-19 vaccine contained equivalent expressed wild-type SARS-CoV-2 spike protein and modified B.1.1.529 spike protein. Only 0.1 mL as one dose was administered by nebulized inhalation through mouth (Table S1). All participants were monitored for 30 min after aerosolized vaccination for immediate adverse reactions (ARs) collection. Adverse events (AEs) were self-reported using dairy card followed by investigators’ review during subject’s onsite visit on Day 28 after the vaccination. Serious adverse events (SAEs) were monitored throughout the study period. Blood samples and nasopharyngeal swabs were collected at 0, 14, and 28 days after the booster for immunogenicity assessments, and we assessed Anti-SARS-CoV-2 RBD specific IgG with ELISA kit manufactured by Vazyme Biotech Co., Ltd (Nanjing, China). A sample was diluted and added in enzyme-labeled plate well. Then, RBD calibrator was added, followed by incubating at 37°C for 30 min. After washing, enzyme-labeled reagent working solution was added and then incubated for another 30 min. We washed and dried the liquid in microplate dewatering centrifuge. Next, the chromogenic solution was added into the enzyme-labeled plate well. OD values of each well were detected by microplate photometer (Thermo Fisher Scientific, Waltham, MA, USA) with 450/630 nm dual wavelength.

ELISpot for cytokines detection

Specific T-cell responses were quantified with an enzyme-linked immunosorbent spot (ELISpot) assay using fresh peripheral blood mononuclear cells (PBMC), which were stimulated with peptide pool (Omicron BA.5) and cultured in an incubator for 20–22 h. First of all, the ELISpot plate was conditioned with medium containing 10% of the serum as used for the cell suspensions, and incubated for 30 min. Then, the cells were incubated in prepared ELISpot plates. With washing for 5 times and diluting the biotin-labeled detection antibody with PBS + 0.5% FBS, the plate was left at room temperature for 2 h. Then, the liquid in the plate was discarded. After diluting and adding Streptavidin-ALP, ELISpot plate was left for 1 h. The substrate (BCIP/NBT-plus) was filtered with a 0.45 μm filter membrane followed by adding 100 μl into each well. After rinsing, reaction was terminated until obvious spots appeared. Spot counting analysis was performed using the ELISpot reader (AID, Strassberg, Germany).

SARS-CoV-2 neutralization assay

Neutralization activity used authentic virus wild-type SARS-CoV-2 and Omicron BA.1 was evaluated using a cytopathic effect (CPE)-based assay as previously described [20]. Serum samples were tested at an initial dilution of 1:8 and then diluted in eight two-fold steps. At the beginning, all samples were mixed with 100 TCID50 virus solution and incubated at 37°C and 5% CO2 for 2 h. Then, 1.2 × 10⁴ Vero E6 cells were added to the virus–serum mixture plate, followed by incubating for 4 days at 37 °C in a humidified environment with 5% CO2 and examining for CPE by the Celigo Imaging Cytometer (Nexcelom Bioscience, Lawrence, MA, USA). The absence or presence of CPE was defined by comparing each well with a positive control (plasma sample showing high SARS-CoV-2 neutralizing activity in infected Vero E6 cells) and a negative control (human serum sample negative for SARS-CoV-2 in ELISA and neutralization assay and Vero E6 cells alone). We defined neutralizing antibody titres less than the detection limit dilution as concentration for 50% of maximal effect (EC50) = 4. We also measured the neutralizing antibody titres against pseudovirus SARS-CoV-2 XBB induced by vaccination using pseudotyped virus neutralization test [21]. With serials dilution, cell control (CC) with only cells and virus control (VC) with pseudovirus SARS-CoV-2 XBB were set. Titres were calculated according to the raw data of relative light units (RLU) from CC, VC, and samples.

Tissue-resident memory T cells

Cell surface staining was used to detect human tissue-resident memory (TRM) T cells. The centrifuged and resuspended isolated nasal mucosa cells were added in the plate, washing, and dead cell staining steps. For the cell surface staining, surface antibody staining solution was prepared with PBS, including PerCP/Cyanine5.5 anti-human CD3 Antibody, Brilliant Violet 510™ anti-human CD8 Antibody, Alexa Fluor® 700 anti-human CD4 Antibody, PE anti-human CD69 Antibody, BV421 anti-human CD103 Antibody and Brilliant Violet 650TM anti-human CD45RO Antibody, and were added in each well and resuspended cells for 20-30 min. After washing and resuspending, the samples are ready for flow cytometer analysis (BECKMAN COULTER, Brea, CA, USA).

Electrochemiluminescence assay for SlgA

Meso Scale Discovery (MSD) electrochemiluminescence assay was used to test secretory IgA (SIgA) level elicited by the vaccines. First step was MSD plate preparation. After using MSD Buffer (20x) to prepare 1x of the working solution and BlokerA solution, the standards were melted and diluted 4-fold in turn with seven standards and one blank. Then, antibodies were diluted and well-prepared. At last, 150 uL GOLD Read Buffer B was added in each well after washing plate, followed by detecting with V-PLEX SARS-CoV-2 Panel 27 (IgA) Kit (MESO SCALE DIAGNOSTICS, LLC., Rockville, MD, USA).

Outcomes

The objective of this clinical trial was to study the safety and immunogenicity profiles of Ad5-nCoVO-IH and Ad5-nCoV/O-IH. The primary safety endpoint was the incidence of ARs within 28 days after the vaccination. The primary endpoint for immunogenicity was the seroconversion rates and geometric mean titres (GMTs) of neutralizing antibody against Omicron variant on day 28 post-vaccination. The secondary endpoints include the SAE occurrence up to six months and cellular immunity assessments on day 14 and day 28 after the booster. Grading of AEs was defined according to the guideline issued by the China State Food and Drug Administration and causality assessment was done by investigators before unblinding. As an exploratory objective, immunogenicity evaluations for mucosal responses in terms of SIgA antibodies and cytokines in TRM T-cells were investigated before vaccination and on days 14 and 28 after the vaccination.

Statistical analysis

The sample size was calculated based on the differential design of the main purpose. Based on the GMT of anti-Omicron antibody 3 months after orally receiving aerosolized Ad5 vectored COVID-19 vaccine in previous clinical trials is 23.47, and the standard deviation is 2.902, the coefficient of variation of GMT is estimated to be 1.45. It is considered that the anti-Omicron antibody 28 days after vaccination is used in this study, we conservatively estimated the coefficient of variation does not exceed 3. Type I error α is set as 0.0125 and the power is 90%. The ratio of GMT of anti-Omicron antibody 28 days after vaccination in the experimental group (Ad5-nCoVO-IH group, Ad5-nCoV/O-IH group) to the control group (Ad5-nCoV-IH group) is assumed to be 4.0 and the superiority margin is set as 1. The participants assigned in experimental groups and control group were 1:1:1. After calculating using PASS16.0 software (Tests for Two Proportions), the sample size of each group was 121. Then, the seroconversion rate of anti-Omicron antibody 3 months after the vaccination of Ad5-nCoV-IH in previous clinical trials was about 85%. Considering the seroconversion rate of anti-Omicron antibody at 28 days after vaccination was used in this study, the estimated seroconversion rate of anti-Omicron antibody of control group was about 87%, and the anti-Omicron antibody in the experimental group was estimated to be 98%. Type I error α is set as 0.0125 and the power is 90%. The participants assigned in experimental groups and control group were 1:1:1. After calculating using PASS16.0 software, the sample size of each group was 137. Taking into account the shedding of participants in the trial, it is planned to include 150 patients per group, for a total of 450 cases. Safety data were analysed by Fisher’s exact test to statistically compare the differences between the groups, and the occurrence time distribution and severity of adverse events were statistically described. The 95% confidence interval of seroconversion rate and antibody positivity rate was calculated by the Clopper–Pearson method, and the chi-square test was used to statistically compare the difference between the groups. The geometric mean and its 95% confidence interval were used to statistically describe the GMT of antibody and its growth factor, which used the variance after logarithmic conversion, and the LSD-t test was used to compare pairs between groups when the difference between groups was statistically significant. The cytokine indexes of each group at different times were statistically analysed by the rank sum test. Statistical analyses were done by a statistician using SAS (version 9.4 or above).

Results

Demographic analysis of subjects

Between September 23 and 26, 2022, 481 participants were recruited and screened for eligibility. 451 participants were enrolled, of whom 150 were randomly assigned to Ad5-nCoVO-IH group, 151 to Ad5-nCoV/O-IH group, and 150 to Ad5-nCoV-IH group. All participants completed the vaccination, and 448 subjects contributed blood samples at 28 days after vaccination (Figure 1). Among the subjects enrolled in this trial, the mean ages of the Ad5-nCoVO-IH, Ad5-nCoV/O-IH and Ad5-nCoV-IH groups were 47.51, 45.93, and 45.22 years old, the mean BMIs were 23.81, 23.99, and 23.55 kg/m², and the interval days between the third and fourth doses were 265.00, 277.16, and 276.33, respectively. Gender distribution and people with medical history were similar among these three groups (Table 1). Overall, the batches were comparable in terms of age, BMI, gender, and the interval days between the third and fourth doses.

Figure 1. Trial profile. Ad5-nCoVO-IH = the B.1.1.529 variant Ad5 vectored COVID-19 vaccine, Ad5-nCoV/O-IH = bivalent Ad5 vectored COVID-19 vaccine, and Ad5-nCoV-IH = Ad5 vectored COVID-19 vaccine.

Table 1. Demographic patterns of the studied population.

	Ad5-nCoVO-IH group	Ad5-nCoV/O-IH group	Ad5-nCoV-IH group
Age, years
18–59 y	112 (74.67%)	117 (77.48%)	120 (80.00%)
≥60 y	38 (25.33%)	34 (22.52%)	30 (20.00%)
Mean (SD)	47.51 (14.47)	45.93 (14.70)	45.22 (14.87)
Sex
Male	61 (40.67%)	62 (41.06%)	60 (40.00%)
Female	89 (59.33%)	89 (58.94%)	90 (60.00%)
BMI
kg/m^2	23.81 (3.41)	23.99 (3.35)	23.55 (3.23)
Previous medical history
Yes	44 (29.33%)	52 (34.44%)	48 (32.00%)
No	106 (70.67%)	99 (65.56%)	102 (68.00%)
Interval between the third and fourth doses, days
Mean (SD)	265.00 (56.24)	277.16 (58.14)	276.33 (61.42)

Safety evaluation

Table 2 tabulates the grading of ARs within 28 days after vaccination in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups. The overall incidence of AR reported from Ad5-nCoVO-IH group, Ad5-nCoV/O-IH group, and Ad5-nCoV-IH group were 24.67%, 18.54%, and 17.33%, with no significant difference among three groups. Dry mouth was a typical symptom reported and is more pronounced in the Ad5-nCoVO-IH group compared to Ad5-nCoV/O-IH and Ad5-nCoV-IH groups, with an incidence of 13(8.67%), 6(3.97%), and 9 (6.00%), respectively. Oropharyngeal pain reported in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were 4.67%, 3.97%, and 4.67%, whilst pharyngeal swelling was only reported in Ad5-nCoVO-IH and Ad5-nCoV/O-IH groups with an incidence of 1.33% and 1.99%, respectively. The most frequently systemic AR was headache, reported by 4.67%, 3.31%, and 4.00% in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups. Fatigue (4.67%) and asthenia (4.00%) were commonly reported by subjects from Ad5-nCoVO-IH group, but the incidence was low in the other groups. Fever was another typical symptom reported, the incidence in Ad5-nCoVO-IH, Ad5-nCoV/O-IH and Ad5-nCoV-IH groups were 5(3.33%), 3(1.99%) and 2 (1.33%). Unsolicited ARs occurred in these three groups were 7.33%, 3.97%, and 5.33%, respectively. Other symptoms that appeared to be related to vaccination can refer to Table 2. Our inter-group comparisons showed that there was no statistically significant difference among these groups for all the symptoms except asthenia exhibited by the subjects. Asthenia in Ad5-nCoVO-IH group (4.00%) showed a significant high incidence compared with Ad5-nCoV/O-IH group (0.66%) and Ad5-nCoV-IH group (0.67%). Also, the severity was mainly mild since most of the symptoms fall under grade 1 and grade 2 categories. Only one grade 3 AR (fever) was reported in one (0.66%) subject in the Ad5-nCoVO-IH group, and no vaccine-related SAE was reported. Taken together, our data showed that the administered inhaled with Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH was well tolerated by the subjects and had a good safety profile due to low number of subjects experiencing undesired side effects.

Table 2. Adverse reactions within 0–28 days after vaccination.

	Ad5-nCoVO-IH group (N = 150)	Ad5-nCoV/O-IH group (N = 151)	Ad5-nCoV-IH group (N = 150)	P-value
	Any (n (%))	Any (n (%))	Any (n (%))
Dysphonia	1 (0.67)	0 (0.00)	0 (0.00)	0.67
Oropharyngeal pain	7 (4.67)	6 (3.97)	7 (4.67)	0.94
Pharyngeal swelling	2 (1.33)	3 (1.99)	0 (0.00)	0.38
Dry mouth	13 (8.67)	6 (3.97)	9 (6.00)	0.24
Stomatitis	1 (0.67)	3 (1.99)	1 (0.67)	0.63
Decreased appetite	1 (0.67)	0 (0.00)	0 (0.00)	0.67
Headache	7 (4.67)	5 (3.31)	6 (4.00)	0.83
Arthralgia	0 (0.00)	0 (0.00)	1 (0.67)	0.67
Myalgia	4 (2.67)	3 (1.99)	5 (3.33)	0.72
Dyspnea	1 (0.67)	0 (0.00)	0 (0.00)	0.22
Cough	4 (2.67)	4 (2.65)	3 (2.00)	1.00
Insomnia	1 (0.67)	0 (0.00)	0 (0.00)	0.67
Hypersensitivity	1 (0.67)	0 (0.00)	0 (0.00)	0.67
Pruritus	1 (0.67)	2 (1.32)	0 (0.00)	0.78
Fever	5 (3.33)	3 (1.99)	2 (1.33)	0.53
Grade 3 fever	0 (0.00)	1 (0.66)	0 (0.00)	–
Asthenia	6 (4.00)	1 (0.66)	1 (0.67)	<0.05*
Fatigue	7 (4.67)	2 (1.32)	1 (0.67)	0.077
Nausea	2 (1.33)	1 (0.66)	1 (0.67)	0.85
Vomiting	0 (0.00)	0 (0.00)	1 (0.67)	0.67
Overall unsolicited AR	11 (7.33)	6 (3.97)	8 (5.33)	0.44
Overall AR	37 (24.67)	28 (18.54)	26 (17.33)	0.24
Grade 3	0 (0.00)	1 (0.66)	0 (0.00)	–

*P < .05 represents significant difference among three groups.

Humoral immunity

Neutralizing antibodies and binding antibodies

At day 14 after vaccination, the seroconversion rates of neutralizing antibodies against BA.1 in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were 4.67%, 5.30%, and 6.00% respectively, followed by significantly increasing to 56.00%, 59.60% and 48.67% at 28 day after vaccination with no significant difference among three groups (Figure 2(A)). For the seroconversion rates of neutralizing antibodies against wild-type SARS-CoV-2, they were 8.22%, 11.49% and 10.07% in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 14 after vaccination, and then dramatically increased to 70.55%, 84.46% and 88.59% at day 28 after vaccination (Figure 2(B)). There was statistically significant difference between Ad5-nCoVO-IH and Ad5-nCoV/O-IH groups (P = .0043), and between Ad5-nCoVO-IH and Ad5-nCoV-IH groups (P = .0001) at day 28 after vaccination. The seroconversion rates of IgG antibodies against SARS-CoV-2 S-RBD were 58.00%, 71.52%, and 72.67% at day 14 after vaccination, and increased to 83.33%, 91.39% and 93.33% in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 28 after vaccination (Figure 2(C)). The difference in the seroconversion rates of IgG antibodies against SARS-CoV-2 S-RBD was statistically significant between Ad5-nCoVO-IH and Ad5-nCoV/O-IH (P = .0353), and between Ad5-nCoVO-IH and Ad5-nCoV-IH (P = .0070) at day 28 post vaccination.

Figure 2. Bar chart illustrates the seroconversion rates of neutralizing antibodies against (A) Omicron BA.1, and (B) wild-type SARS-CoV-2 as well as (C) IgG antibodies against SARS-CoV-2 S-RBD between the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 14 and day 28 (post-vaccination). The inter-group comparison was performed using LSD-t method. *denotes statistically significant between the groups (P < .05).

The GMTs of neutralizing antibodies against Omicron BA.1 (live virus) in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were 4.13, 4.00, and 4.10, respectively, before vaccination, and 5.10, 4.91 and 5.01 at 14 days after vaccination, and had drastically spike to 31.01, 43.70 and 29.25 at 28 days after vaccination (Figure 3(A)). The difference in the GMTs of neutralizing antibodies against Omicron BA.1 was statistically significant between Ad5-nCoV/O-IH and Ad5-nCoV-IH at day 28 post vaccination (P = .0238). The GMTs of neutralizing antibodies against wild-type SARS-CoV-2 were similar before vaccination and drastically increased to 94.34, 222.41, and 267.83 at day 28 after boosted with Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH. Ad5-nCoV-IH and Ad5-nCoV/O-IH inducted significantly higher neutralizing antibodies than that induced by Ad5-nCoVO-IH vaccine (Figure 3(B)). We also performed IgG antibodies response against SARS-CoV-2 S-RBD at day 0 (pre vaccination), at day 14, and day 28 post vaccination (Figure 3(C)). The GMTs of IgG antibodies against SARS-CoV-2 S-RBD in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were greatly increased from 61.03, 50.60 and 53.20, respectively, before vaccination, to 518.56, 935.09 and 1088.66 at 14 days after vaccination, and were continually rising to 1,913.43, 3,052.98 and 3,428.50 at 28 days after vaccination. Statistical analysis unveiled the significant differences in the GMTs of IgG antibodies between the Ad5-nCoVO-IH and Ad5-nCoV/O-IH groups and between the Ad5-nCoVO-IH and Ad5-nCoV-IH groups at day 14 and day 28 post vaccination. At day 28 after vaccination, the GMTs of pseudovirus-neutralizing antibodies against Omicron XBB were 41.46, 32.96, and 29.95 in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups, respectively, with no significant difference among three groups (Figure 3(D)).

Figure 3. Bar chart illustrates the GMTs of neutralizing antibodies against (A) authentic virus Omicron BA.1, and (B) authentic virus wild-type SARS-CoV-2, and pseudovirus XBB as well as (D) IgG antibodies against SARS-CoV-2 S-RBD between the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 0 (pre-vaccination), day 14 and day 28 (post-vaccination). The inter-group comparison was performed using LSD-t method. *denotes statistically significant between the groups (P < .05).

Overall, our study showed that all groups reported a high escalation in the neutralizing antibodies production against Omicron subvariants and wild-type SARS-CoV-2 at day 28 post vaccination. Nonetheless, Ad5-nCoV/O-IH showed highest neutralizing antibodies production against the Omicron BA.1.

Cellular immunity analysis

As shown in Figure 4, significant increment of IFN-γ and IL-2 were detected in all treatment groups and Ad5-nCoVO-IH group showed more superior of these cytokines production than Ad5-nCoV/O-IH and Ad5-nCoV-IH groups at day 14 post vaccination.

Figure 4. Bar chart showing the antibody level of IFN-γ and IL-2 in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 0 (pre vaccination) and day 14 post vaccination. Results were mean ± SD of 20 subjects in each group.

All the groups showed high positivity rates of IFN-γ and IL-2 at day 14 post vaccination. The IFN-γ positive rates in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were 90.00%, 95.00%, and 75.00%, and the IL-2 were 95.00%, 95.00% and 90.00%, respectively. However, the differences in positive rates of these cytokines at day 14 post vaccination between Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups were not statistically significant (Table S2).

Nasopharyngeal swab SIgA analysis

Given that the inhaled vaccine directly reaches the respiratory tract and lungs and interact with the immune cells located in the mucosa creating immune responses at site, as well as the evidence of increased SIgA antibodies in BAL sample from animal model [13], we further explored the mucosal immunity triggered by the inhaled vaccine. Thus, participants were asked to provide nasopharyngeal swab samples for SIgA antibodies assessment. We studied the GMTs of nasal SIgA antibodies against the Spike of wild type SARS-CoV-2 and 6 VOCs or their mutants (beta, delta, BA.2, BA.2 + L452M, BA.2 + L452R, BA.2.12.1, BA.3, BA.4 and BA.5) in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH and Ad5-nCoV-IH groups. As shown in Figure 5, increased in SIgA antibodies response for different Spikes were reported at day 14 and day 28 post vaccination compared to day 0 pre vaccination in all groups. Besides, the antibodies showed high percentages of more than two-fold increase at day 28 post vaccination, ranging from 55% to 95% (Table S3). Except for the low response to BA.3 Spike, the SIgA antibodies were good in blocking the combination of ACE2 and the Spike of SARS-CoV-2 and its variants, including BA.5. Furthermore, no difference was found in terms of a specific virus/variant among these treatment groups, suggesting a broader immune effects offered by SIgA.

Figure 5. Bar chart showing the GMT of nasal SIgA antibodies between the Ad5-nCoVO-IH, Ad5-nCoV/O-IH and Ad5-nCoV-IH groups against SARS-CoV-2 B.1.617.2, B.1.351, BA.2, BA.2 + L452M, BA.2 + L452R, BA.2.12.1, BA.3, BA.4, BA.5, and SARS-CoV-2 spike, before vaccination, 14 days post-vaccination and 28 days post-vaccination. Results were presented in mean ± SD of 20 subjects in Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 0 (pre-vaccination), day 14 post vaccination, and day 28 post vaccination. *denotes statistically significant between the 14 or 28 days post-vaccination and before vaccination (P < .001).

Mucosal immunity resident memory T-cells analysis

We also performed an analysis on the expression of CD103 in tissue-resident memory (TRM) T-cells and the results are presented in Figure 6. CD4⁺ CD45RO⁺ CD69⁺ CD103⁺ and CD4⁺ CD45RO⁺ CD69⁺ CD103^- levels showed an increasing pattern starting from day 0 to day 14 and day 28 post vaccination. The levels of CD8⁺ CD45RO⁺ CD69⁺ CD103⁺ were greatly increased in all groups at day 14 and day 28 post vaccination when compared to that before vaccination. The highest increment was observed in Ad5-nCoV/O-IH group, followed by Ad5-nCoV-IH group and lastly Ad5-nCoVO-IH group. For CD8⁺ CD45RO⁺ CD69⁺ CD103^-, the levels were increased in the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 14 and day 28 post vaccination, with a slight reduction was observed in the Ad5-nCoVO-IH and Ad5-nCoV-IH groups at day 28 post vaccination when compared to day 0 pre vaccination. These results may demonstrate that the inhaled vaccine well triggered mucosal immune responses. No statistical difference was found among these three vaccine groups, while the increased expression of these markers suggests that the aerosolized adenovirus type-5 vector-based COVID-19 vaccines are able to activate T-cell responses at mucosal level and then trigger systemic immune response throughout the body.

Figure 6. Scatter plot showing the mucosal immunity TRM analysis between the Ad5-nCoVO-IH, Ad5-nCoV/O-IH, and Ad5-nCoV-IH groups at day 0 (pre vaccination), day 14 and day 28 post vaccination. Results were presented in mean ± SD of 20 subjects in each group for all visits. The inter-group comparison of mucosal immunity TRM analysis was performed using Wilcoxon test.

Discussion

This is the first clinical trial that evaluated the safety and immunogenicity of the aerosolized bivalent Ad5 vector-based COVID-19 vaccine (Ad5-nCoV/O-IH). Our study showed that Ad5-nCoVO-IH and Ad5-nCoV/O-IH had comparable safety profile like Ad5-nCoV-IH group (as shown in Table 2). The common local ARs in all groups include oropharyngeal pain, stomatitis, and dry mouth while the common systemic ARs in all groups include myalgia, cough, headache, and fever. In addition, the most common unsolicited AR was dizziness, reported in Ad5-nCoVO-IH and Ad5-nCoV/O-IH groups. Besides, this study also showed that the aerosolized vaccines did not compromise the health of the subjects as low percentage of ARs (below 10%) were reported in Ad5-nCoV/O-IH, Ad5-nCoVO-IH, or Ad5-nCoV-IH groups. Majority of the reported ARs were mild to moderate, except for one (0.66%) grade 3 fever was reported in Ad5-nCoV/O-IH. This further tells us that the safety of the vaccine was well tolerated by the subjects in all groups. This is in agreement with previous study which reported that the heterologous boost immunization with inhaled Ad5 vector-based COVID-19 vaccine was safe and had lower adverse reactions [22].

Neutralizing antibody responses have been used to inform COVID-19 vaccine efficacy [23,24]. This study showed that the aerosolized bivalent vaccine can induce higher neutralizing antibody response against Omicron BA.1 28 days after the booster dose, indicating Ad5-nCoV/O-IH might be effective to be used as a booster vaccine to protect against Omicron variants. The significantly high neutralizing antibody responses against wild-type SARS-CoV-2 and S-RBD IgG antibodies in Ad5-nCoV-IH and Ad5-nCoV/O-IH groups suggest that they have a predilection towards the production of antibody against wild-type virus. Mechanisms of enhanced antibody responses with bivalent vaccines have yet to be unravelled but could be attributed to the generation of new memory immune responses [24]. Notably, at day 28 after vaccination, the seroconversion rate of neutralizing antibodies against wild-type SARS-CoV-2 in Ad5-nCoV-IH group was significantly higher than that in Ad5-nCoVO-IH group, while the significant difference in the seroconversion rate of neutralizing antibodies against BA.1 was not shown between these two groups. Moreover, in the Ad5-nCoVO-IH group, the neutralizing antibody responses against wild-type SARS-CoV-2 at day 28 after vaccination were 9.01-fold higher than those before vaccination (94.34 vs 10.47), while the neutralizing antibody responses against BA.1 at day 28 after vaccination were 7.53-fold higher than those before vaccination (31.10 vs 4.13). As the vaccination history of participants was three doses of inactivated COVID-19 vaccine without infection before, these performances of Ad5 containing Omicron-boosters should be interfered by immune imprinting. Since boosting with a variant mutated from WT would majorly recall memory B cells induced by WT vaccination or infection and hider the humoral immunity against new and emerging variants from de novo generation of variant-specific B cells [25]. Considering Ad5-nCoV/O-IH showed a significant higher GMTs of neutralizing antibodies against Omicron BA.1 compared with Ad5-nCoV-IH at day 28 post vaccination, monovalent Omicron COVID-19 vaccination could be a development direction, including XBB vaccines against current prevalent strain. However, as the results from animal and human trials, the influence of immune imprinting could be diminished by repeat vaccination of COVID-19 variants vaccines [25,26]. In addition, there was a study indicated the immunogenicity effect of boosting with inactivated vaccine, aerosolized Ad5-nCoV vaccine and mRNA vaccine against omicron variants including XBB.1.5, BQ.1, BQ.1.1, BF.7 and BA.2.75.2. As the results shown, heterologous booster with an aerosolized Ad5-nCoV vaccine after two-dose priming with inactivated vaccine seemed to have a comparable effect compared with other vaccination [27]. These findings also illustrate aerosolized vaccine can be a good strategy for future SARS-CoV-2 booster doses.

We also first reported the nasal SIgA antibody response against the spike of wild-type SARS-CoV-2 and VOCs including beta, delta, and omicron variants induced by aerosolized Ad5 vector-based COVID-19 vaccines. SIgA is the main effector of mucosal immunity, preventing viral invasion by removing respiratory pathogens before they cross the mucosal barrier [28]. In SARS-CoV-2 infection, IgA antibodies that attach to the virus are instantly produced, even before IgG antibodies [29–32]. However, it remains obscure to what extent inhaled Ad5 vectored COVID-19 vaccine could induce SIgA antibody responses against the virus. One study had shown that intramuscular mRNA vaccination is able to induce weak mucosal SIgA response and is impacted by pre-existing immunity [33] and is consistent with other studies that have shown that SIgA antibodies in saliva can be induced by intramuscular mRNA vaccination under certain circumstances [34,35]. Our study showed that all groups (Ad5-nCoV/O-IH, Ad5-nCoVO-IH, or Ad5-nCoV-IH groups) were able to induce substantial amount of SIgA antibodies against SARS-CoV-2 and VOCs post vaccination. The highest SIgA antibody response was against wild-type SARS-CoV-2, the lowest response was detected against BA.3, and while against other variants are comparable. Additionally, there is a broader and relatively high percentage of SIgA antibodies more than two-fold increase against original coronavirus and variants, which also showed an increase from day 14 to day 28 post vaccination. Overall, these results indicate that inhaled Ad5 vector-based COVID-19 vaccine could result in better nasal SIgA production against wild-type SARS-CoV-2 and VOCs. Based on our findings, we believe that the inhaled Ad5 vectored COVID-19 vaccines may have utility in preventing infection by eliciting robust mucosal immunity against wild-type SARS-CoV-2 and VOCs.

This study also characterizes the expression of TRM T-cells, defined by the co-expression of CD69 and CD103 [36]. We observed an overall increment of CD4⁺ and CD8⁺ TRM cells in all groups post vaccination, except for CD103^-, a slight reduction was observed in the Ad5-nCoVO-IH and Ad5-nCoV-IH groups at day 28 post vaccination when compared to that before vaccination. The findings here are encouraging since increasing evidence suggests that TRM T-cells confer protection during different pathogenic infections [37,38]. Evidence showed that increased frequencies of CD4⁺ cells in the airway in humans are linked with surviving severe disease of SARS-CoV-2 infection [39]. Also, enormous amounts of memory CD8⁺ T-cells reside in the tissues attributed to localized SARS-CoV-2–specific immune responses [40]. Further, the presence of preexisting CD8⁺ TRM cells in unexposed individuals could potentially initiate rapid sentinel immune responses against SARS-CoV-2 [40]. In a review, Hirahara et al. reported the crucial roles of CD8⁺ TRM cells in the local control of viral infection [41]. We hypothesized that increase expression of CD4⁺ and CD8⁺ TRM cells will induce the secretion of immune mediators, for example, specific cytokines and chemokines that will protect the host from SARS-CoV-2 infection as well as facilitate an immediate immune response against re-exposure to SARS-CoV-2. Also, a comprehensive understanding of the function of CD4⁺ and CD8⁺ TRM cells is pertinent for the development of effective vaccination against SARS-CoV-2. Therefore, the results presented here highlight that targeted mucosal immunization could be promising against wild-type SARS-CoV-2 and VOCs.

During the study, Ad5-nCoV/O-IH showed a higher neutralizing capability against Omicron BA.1 compared with Ad5-nCoV-IH, while Ad5-nCoVO-IH groups had the lowest neutralizing antibodies against wild-type SARS-CoV-2 among three groups. Considering mucosal immunity, there was no significant difference in the detection indicators among three groups. On the one hand, it was thought that all three vaccines were aerosolized inhalation vaccines, so the way they stimulated mucosal immunity was same. On the other hand, the mechanism of mucosal immunity induced through oral administration is not clear enough and the validated assays for nasopharyngeal swab evaluation were need to be further explored [42]. Furthermore, Ad5-nCoV/O-IH could provide a broad-spectrum immune response for against wild-type SARS-CoV-2 and Omicron variants.

In conclusion, this study showed that Ad5-nCoV/O-IH is safe and highly immunogenic and could be an effective strategy to control the expected spike in wild-type and Omicron-induced infections in the COVID-19 pandemic. Since SARS-CoV-2 infection begins at the respiratory tract, it is critical to target the virus at its entry point. There are some limitations in the study. For instance, the vaccines used in the study were all aerosolized vaccines, so there is a lack of comparison of safety and immunogenicity especially the mucosal immunity induced by different administration routes. In addition, with the rapid mutation of SARS-CoV-2, the new and emerging variants including BA.4/5, BQ.1.1, and so on, became more popular over the world. Although we detected the pseudovirus neutralizing antibodies against XBB and SIgA Against BA.4/5, our results may not be fully extrapolated to other untested strains. Besides, inconsistency occurred during different experiment tests. Because the primary immunogenicity endpoints of the study were GMT and seroconversion rate of anti-Omicron-specific neutralizing antibodies, neutralizing antibodies against authentic virus SARS-CoV-2 original strain and BA.1 were detected. However, PBMCs were stimulated with BA.5 instead of BA.1 peptide pool during ELISpot analysis, in which BA.5 was considered more prevalent followed by mutation of SARS-CoV-2. While cellular immunity and mucosal immunity were exploratory endpoints, the effect of vaccines on kinds of variants was decided. Overall, our findings support that aerosol vaccination with Ad5 vectored COVID-19 vaccine nor is bivalent Ad5 vectored COVID-19 vaccine a promising approach to deliver COVID-19 vaccines due to elicitation of robust humoral, cellular, and mucosal immune responses. As the trend showed, the immune response was increased at 28 days after vaccination compared with before, and we will continue to conduct immunogenicity persistence further study.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was supported by Chongqing Health Commission and Chongqing Science and Technology Commission [Grant 2020FYYX040], Study on SARS-CoV-2 vaccine in elderly people over 60 years old in Chongqing [Grant 2022ZDXM030], and the First Batch of Key Disciplines on Public Health in Chongqing, and Tianjin Biomedical Science and Technology Major Project [Project number 21ZXSYSY00040].