Comparison of Immunogenicity and Reactogenicity of Five Primary Series of COVID-19 Vaccine Regimens against Circulating SARS-CoV-2 Variants of Concern among Healthy Thai Populations

To compare immunogenicity and reactogenicity of five COVID-19 vaccine regimens against wild-type SARS-CoV-2 and variants of concern (VoCs) among Thai populations, a prospective cohort study was conducted among healthy participants aged ≥18 years who had never been infected with COVID-19 and were scheduled to get one of the five primary series of COVID-19 vaccine regimens, including CoronaVac/CoronaVac, AZD1222/AZD1222, CoronaVac/AZD1222, AZD1222/BNT162b2, and BNT162b2/BNT162b2. Anti-receptor binding domain (anti-RBD-WT) IgG and neutralizing antibody (NAb-WT) against wild-type SARS-CoV-2 were measured at pre-prime, post-prime, and post-boost visits. NAb against VoCs (NAb-Alpha, NAb-Beta, NAb-Delta, and NAb-Omicron) were assessed at the post-boost visit. Adverse events (AEs) following vaccination were recorded. A total of 901 participants (CoronaVac/CoronaVac: 332, AZD1222/AZD1222: 221, CoronaVac/AZD1222: 110, AZD1222/BNT162b2: 128, and BNT162b2/BNT162b2: 110) were enrolled. Anti-RBD-WT IgG and NAb-WT levels increased substantially after each vaccine dose. At the post-boost visit, BNT162b2/BNT162b2 induced the highest GMC of anti-RBD-WT IgG level (1698 BAU/mL), whereas AZD1222/BNT162b2 induced the highest median NAb-WT level (99% inhibition). NAb levels against VoCs, particularly the Omicron strain, were markedly attenuated for all vaccine regimens (p < 0.001). Overall, no serious AEs following vaccination were observed. All five primary series of COVID-19 vaccine regimens were well-tolerated and elicited robust antibody responses against wild-type SARS-CoV-2 but had attenuated responses against VoCs, particularly the Omicron strain, among healthy Thai populations.


Introduction
In the light of the global pandemic of coronavirus disease in 2019 (COVID- 19), vaccination against the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the most efficient and vital strategy to protect against infection and prevent severe disease, hospitalization, and death [1][2][3]. As of January 2023, 50 vaccines based on different platforms, including inactivated whole viruses, viral vectors, nucleic acids, and protein subunits, with different efficacies, have been authorized or licensed for use in at least one country, of which 11 have been granted emergency use listing by the World Health Organization (WHO) [4]. To date, a total of 13.3 billion doses of the COVID-19 vaccine have been administered worldwide, and 1.2 million are administered each day. Overall, 69% of the world's population has received at least one dose of the COVID-19 vaccine, whereas only 26% of people in low-income countries have received at least one dose [5].
During the initial phase of the COVID-19 vaccination rollout, most of the approved vaccines followed a homologous prime-boost regimen, in which the same type of vaccine was administered for a priming and a booster dose. However, regarding the concerns of serious adverse events following immunization (AEFIs), waning of vaccine-elicited immunity, new emerging variants of concern (VoCs), and intermittent vaccine supply shortages, a heterologous prime-boost vaccine schedule, which is the use of different types of vaccines for a priming and a booster dose, has gained increasing interest as an alternative vaccine regimen, particularly for low-and middle-income countries. There is growing evidence that this vaccine regimen could provide robust immune responses with safe reactogenicity profiles [6][7][8].
Antibodies have been established as a clear correlate of protection against SARS-CoV-2 infection after vaccination [9,10]. However, the immunogenicity of each COVID-19 vaccine regimen varies widely across studies due to variation in the study population, study setting, vaccine dosage and immunization schedule, laboratory measurement of vaccine-elicited antibodies, and predominant SARS-CoV-2 variants during the study period, which caused challenges in comparing the immunogenicity across vaccine regimens. To date, few headto-head comparative studies have been conducted to assess antibody responses to diverse vaccine regimens [11]. Additionally, the fast-track evaluation and approval of currently available COVID-19 vaccines have made it necessary to demonstrate the real-world safety and reactogenicity of the vaccines [12].
In Thailand, inactivated whole virus (e.g., CoronaVac) and adenovirus-vectored vaccines (e.g., AZD1222) were among the first vaccine platforms authorized for the general population during the early phase of the COVID-19 pandemic, followed by mRNA vaccines (e.g., BNT162b2). The recommendations for a primary series of the COVID-19 vaccine regimen have been continuously updated by the Ministry of Public Health of Thailand (MOPH, Thailand), according to the availability of vaccine supplies in the country [13]. As of 2 December 2022, approximately 83% of the Thai population have received at least one dose, and 78% have completed a two-dose primary vaccine course [14]. This study aimed to compare the immunogenicity and safety profiles of five primary series of COVID-19 vaccine regimens against wild-type SARS-CoV-2 and other important circulating VoCs among healthy Thai populations.

Study Design and Study Population
This is a single-center prospective cohort study carried out from June 2021 to January 2022 at the Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand. According to the WHO report, there had been 2290 to 23,418 confirmed COVID-19 cases with 9 to 301 COVID-19-related deaths per day in Thailand during the study period [15]. Healthy participants aged 18 years and older who had neither a confirmed diagnosis of COVID-19 nor previously received any vaccines against SARS-CoV-2 and were scheduled to get one of the five primary series of COVID-19 vaccine regimens, including Coron-aVac/CoronaVac, AZD1222/AZD1222, CoronaVac/AZD1222, AZD1222/BNT162b2, or BNT162b2/BNT162b2, were enrolled. Participants with a history of vaccine-associated hypersensitivity, unstable or uncontrolled comorbidities, an immunocompromised state, receiving immunosuppressive agents, or pregnancy were excluded. The study was approved by the Research Ethics Committee of the Faculty of Medicine, Chiang Mai University (approval numbers: 187/2021; 232/2021; 344/2021; and 413/2021). All participants provided written informed consent prior to study enrollment.

Data Collection and Safety Assessment
At enrollment, participant characteristics and the history of administration of other vaccines within the past 6 months were collected. Any immediate adverse events (AEs) observed within 30 min following vaccination were recorded. Additionally, all participants were instructed to self-assess and self-report any AEs, both local and systemic, using an electronic vaccine diary for seven days after each dose of vaccine. The solicited local AEs included pain, erythema, and swelling at the injection site, whereas the solicited systemic AEs included fatigue, headache, fever, paresthesia, and dizziness.

Blood Collection
Blood samples were collected at the pre-prime (before the priming dose), post-prime (3-4 weeks after the priming dose, depending on the vaccine regimen), and post-boost visits (4 weeks after the booster dose). Sera were extracted, aliquoted, and stored at −70 • C until the laboratory analyses.
2.5. Serological Testing 2.5.1. Chemiluminescent Microparticle Immunoassay for Wild-Type SARS-CoV-2 RBD IgG Antibody All sera were analyzed for immunoglobulin G (IgG) antibodies to the receptor-binding domain of the S1 subunit of wild-type SARS-CoV-2 spike protein (anti-RBD-WT IgG) using the SARS-CoV-2 IgG II Quant assay (kit catalog number: 06S6122) on the Alinity i System (Abbott Diagnostics, Chicago, IL, USA), according to the manufacturer's instructions. The chemiluminescent microparticle immunoassay (CMIA) technique was used in this automated immunoassay for the quantitative determination of anti-RBD-WT IgG level. In brief, the participant's serum, SARS-CoV-2 antigen-coated paramagnetic microparticles, and assay diluent were combined and incubated. The anti-RBD-WT IgG presenting in the participant's serum was bound to the SARS-CoV-2 antigen-coated microparticles. Following the addition of an anti-human IgG acridinium-labeled conjugate, the resulting chemiluminescent reaction was measured as a relative light unit (RLU). The amount of anti-RBD-WT IgG for each participant was determined by comparing the chemiluminescent RLU to the IgG II calibrator RLU. The unit of anti-RBD-WT IgG antibody (arbitrary unit [AU]/mL) was converted to binding antibody units (BAU)/mL using the equation suggested by the manufacturer (BAU/mL = 0.142 * AU/mL). The cut-off value of ≥50 AU/mL or ≥7.1 BAU/mL was considered seropositive.

Surrogate Virus Neutralization Test for Wild-Type SARS-CoV-2
Neutralizing antibodies against wild-type SARS-CoV-2 (NAb-WT) were measured with the surrogate virus neutralization test (sVNT) in enzyme-linked immunosorbent assay (ELISA) format, using the SARS-CoV-2 NeutraLISA (Euroimmun AG, Lübeck, Germany). The result was shown in percentage of inhibition (% inhibition), and the cut-off value of ≥35% was defined as detectable neutralizing activity.

Statistical Analysis
Participant characteristics and AEFIs within seven days after each dose of vaccine were presented as number (percentage) and median (interquartile range [IQR]) for categorical and continuous variables, respectively. The comparisons of such parameters across five COVID-19 vaccine regimens were performed using the Chi-squared test or Fisher's exact test (if less than five observations) for categorical variables, and Wilcoxon rank-sum test for continuous variables.
Anti-RBD-WT IgG level was presented as geometric mean concentration (GMC) and 95% confidence interval (95% CI), and NAb titer against each SARS-CoV-2 strain (NAb-WT, NAb-Alpha, NAb-Beta, NAb-Delta, and NAb-Omicron) was reported as median and IQR. The comparisons of GMCs of anti-RBD-WT IgG between study visits within each vaccine regimen were performed using the mixed-effects log-linear regression model, whereas the comparisons between indicated vaccine regimens were conducted using the log-linear regression model. The magnitude of the difference was summarized with a geometric mean ratio (GMR) and 95% CI. In addition, the comparisons of NAb between study visits and between SARS-CoV-2 strains within each vaccine regimen were conducted using the Wilcoxon signed-rank test, while the comparisons between the indicated vaccine regimens were performed using the median regression analysis. The magnitudes of the differences were summarized with a median difference and an IQR. For the proportion of seropositive participants, the comparisons between study visits and between SARS-CoV-2 strains within each vaccine regimen were conducted using the McNemar's test, whereas the comparisons between indicated vaccine regimens were performed using the Chi-squared test. The magnitudes of the differences were summarized with a proportion difference and 95% CI. Furthermore, within each vaccine regimen, the log-linear regression analyses and the Wilcoxon rank-sum tests were performed to compare anti-RBD-WT IgG levels as well as NAb levels against wild-type SARS-CoV-2 and other VoCs between adult and elderly participants, respectively. All statistical analyses were conducted using Stata statistical software, version 17 (StataCorp LP, College Station, TX, USA), and GraphPad Prism, version 9.0 (GraphPad, San Diego, CA, USA). A two-sided p < 0.05 was considered statistically significant.

Adverse Events Following a Primary Series of COVID-19 Vaccination
No immediate AEs were observed within 30 minutes following each vaccine dose in all regimens. During the seven days after a priming dose, the overall incidence of solicited AEs was 57%, 81%, 57%, 90%, and 94% among participants receiving Coron-aVac/CoronaVac, AZD1222/AZD1222, CoronaVac/AZD1222, AZD1222/BNT162b2, and BNT162b2/ BNT162b2 vaccine regimens, respectively. The most common local AE was pain at the injection site, which was greatest among participants in the BNT162b2/BNT162b2 group (93%). Fatigue was the most common systemic AE, and the incidence was greatest among those in the AZD1222/BNT162b2 group (61%) ( Table 5). After a booster dose, the overall incidence of solicited AEs was 57%, 71%, 76%, 68%, and 96% among participants receiving CoronaVac/CoronaVac, AZD1222/AZD1222, CoronaVac/AZD1222, AZD1222/BNT162b2, and BNT162b2/BNT162b2 vaccine regimens, respectively. Similar to a priming dose, pain at the injection site was the most common local AE and was greatest among participants in the BNT162b2/BNT162b2 group (93%); whereas fatigue was the most common systemic AE and was greatest among participants in the CoronaVac/AZD1222 group (45%) ( Table 5). All reported AEs were mild. The majority occurred within the first three days following vaccination and spontaneously resolved within a few days without requiring any specific treatment. No serious AEs were observed in this study. Table 4. The median of neutralizing antibodies, based on the in-house surrogate virus neutralization test, against wild-type and other circulating variants of concern of SARS-CoV-2 following the completion of a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen.    VoC, variant of concern; a The Wilcoxon signed-rank test was performed to compare the medians of neutralizing antibody, based on the in-house surrogate virus neutralization test, between wild-type SARS-CoV-2 and the indicated variant of concern within each COVID-19 vaccine regimen; * Indicates a significant difference (p < 0.05) of the median of neutralizing antibodies, based on the in-house surrogate virus neutralization test, against wild-type and other circulating variants of concern of SARS-CoV-2 after the completion of the primary series of an indicated COVID-19 vaccination compared with the homologous CoronaVac vaccine regimen, evaluated by the median regression analysis; † Indicates a significant difference (p < 0.05) of the median of neutralizing antibodies, based on the in-house surrogate virus neutralization test, against wild-type and other circulating variants of concern of SARS-CoV-2 after the completion of the primary series of an indicated COVID-19 vaccination compared with the homologous AZD1222 vaccine regimen, evaluated by the median regression analysis. Figure 3. Neutralizing antibody, based on in-house surrogate virus neutralization test, against wildtype SARS-CoV-2 and other circulating variants of concern following the completion of a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen. Data points are the reciprocals of individual participants. The median values of neutralizing antibody (NAb; in-house surrogate virus neutralization test) and the corresponding interquartile ranges (IQR) against wildtype SARS-CoV-2 and other circulating variants of concern following the completion of a primary series of COVID-19 vaccination are shown as black solid lines. Black dotted lines indicate the cutoff values. The Wilcoxon signed-rank test was performed to compare the medians of NAb between wild-type SARS-CoV-2 and the indicated variant of concern within each COVID-19 vaccine regimen, and the results are shown in the area above the figure. The median regression analysis was performed to compare the medians of NAb to wild-type SARS-CoV-2 or the indicated variant of concern between COVID-19 vaccine regimens, and the results are shown in the table below the

Discussion
In healthy Thai populations, all five primary series of COVID-19 vaccine regimens, including homologous CoronaVac, AZD1222, and BNT162b2, as well as heterologous Coro-naVac/AZD1222 and AZD1222/BNT162b2, were well-tolerated and highly immunogenic against wild-type SARS-CoV-2. However, the immunogenicity, both anti-RBD IgG and NAb, against SARS-CoV-2 variants, particularly the Omicron strain, was markedly attenuated. Patterns of vaccine-elicited antibody responses were similar for adult and elderly participants, but higher levels were obtained in adults at the post-boost visit. This study provides important information regarding the immunogenicity and safety profiles of several primary series of COVID-19 vaccine regimens, which could help guide the appropriate primary and alternative vaccine regimens for countries with limited vaccine supplies and restricted access to some vaccine platforms.
In this study, homologous BNT162b2 generated a robust and rapid humoral immune response to wild-type SARS-CoV-2. Compared with the other vaccine regimens, it induced the highest anti-RBD-WT IgG level with a 100% seroconversion at 4 weeks following an initial dose and a 9-fold increase in antibody level following a second dose of vaccine. Similar to a population-based prospective cohort study in Northern Cyprus, the homologous BNT162b2 vaccine regimen induced the highest anti-RBD-WT IgG level and seropositivity in adult (aged < 60 years) and elderly participants (aged ≥ 60 years), followed by the homologous AZD1222 and CoronaVac regimens [17]. Although, in this study, heterologous AZD1222/BNT162b2 elicited a significantly lower anti-RBD-WT IgG level compared to a homologous BNT162b2 regimen, the antibody level at the post-boost visit was considerably higher compared to a homologous AZD1222 regimen, which was consistent with earlier studies in European countries [7,18]. We also noted that the heterologous Coro-naVac/AZD1222 vaccine regimen induced significantly higher binding antibodies than either homologous CoronaVac or AZD1222 vaccination, corresponding to previous studies in Thailand [19,20].
Focusing on functional antibodies, the highest vaccine-elicited NAb-WT in this study was observed in the heterologous AZD1222/BNT162b2 vaccine group. Consistent with other findings, vaccination priming with AZD1222 and boosting with BNT162b2 induced a stronger NAb-WT than homologous AZD1222 or BNT162b2 [21,22]. In addition, we noted that heterologous CoronaVac/AZD1222 induced a stronger NAb-WT response than homologous CoronaVac and AZD1222 vaccination, which was similar to the previous study [20]. To date, the mechanism of robust immunogenicity induced by heterologous COVID-19 vaccination is not clearly understood. The possible explanations might be due to the distinct types of host immunity against a specific antigen activated by different vaccine platforms [23] and the curtailment of anti-vector immunity caused by subsequent AZD1222 vaccination [24].
Phylogenetic analysis of SARS-CoV-2 genomes found a substantial number of mutations in the RBD region of SARS-CoV-2 variants. Such mutations considerably impair the binding affinity of vaccine-induced antibodies, resulting in immune evasion of the variants [25,26]. Consistent with previous findings [27][28][29], this study demonstrated that vaccine-elicited NAb levels against important circulating VoCs, particularly the Omicron strain (reduced by 4 to 10-fold), were substantially attenuated for all vaccine regimens compared with those of wild-type SARS-CoV-2. Since all studied COVID-19 vaccine regimens are designed to generate immunity against the spike protein of the wild-type virus, neutralizing capacities against SARS-CoV-2 variants might be impaired [25]. This highlights the need for new COVID-19 booster vaccines, e.g., a bivalent mRNA vaccine that contains mRNA encoding wild-type and Omicron spike proteins to counter the emergence of vaccine-resistant mutations [30].
This study illustrated the poorer vaccine-induced humoral immune responses among elderly participants receiving the homologous AZD1222, and the heterologous Coron-aVac/AZD1222 and AZD1222/BNT162b2 vaccine regimens, compared with adult participants. The weakened antibody responses to COVID-19 vaccines in an aging population were previously demonstrated in a number of studies with several vaccine regimens [31][32][33]. The underlying mechanism is immune senescence, which results in a decline in adaptive immunity with increasing age [34]. Thus, elderly populations remain vulnerable to COVID-19 despite primary vaccination, and a booster vaccine is necessary.
All five COVID-19 vaccine regimens are safe and well-tolerated among our participants. Similar to previous studies, transient pain at the injection site and fatigue were the primary solicited local and systemic reactions [35,36]. We noted that an injection site pain was related to the BNT162b2 vaccine, which might be attributed to temporal fasciitis of the deltoid muscle as there is evidence showing a significant increase in fascia thickness without intramuscular echogenicity change in ultrasound images among vaccine recipients [37]. Additionally, we found that fatigue was associated with the AZD1222 vaccine, which might be due to a mild and short-lived increase in circulating inflammatory cytokines in vaccinees [38]. Although no immediate or serious AEs were observed throughout the study period, with a relatively small study cohort, we had limited ability to ascertain rare serious AEs for each vaccine regimen.
This study contains some strengths. We were able to conduct a population-based, head-to-head comparative study to evaluate humoral immunogenicity and reactogenicity of five primary series of COVID-19 vaccine regimens against wild-type SARS-CoV-2 and important circulating VoCs, including the Omicron strain. Nevertheless, there are some limitations. Firstly, since this study is not an RCT, several potential unknown and uncontrollable confounding factors that could bias the study outcomes might not be appropriately adjusted. The studied vaccine regimens relied on vaccine supplies from the Thai government, and eligible populations were determined by the recommendations of the MOPH of Thailand during the study period; therefore, some specific vaccines, e.g., mRNA1273 and Sinopharm BBIBP, as well as some specific populations, e.g., elderly participants in homologous CoronaVac and BNT162b2 vaccine groups, were not evaluated. In addition, the significant difference in age of participants across the five vaccine groups might contribute to the differences in humoral immune responses to vaccination demonstrated in this study. Moreover, while robust antibody responses following primary vaccination have been demonstrated, the durability and trajectory patterns are needed to be monitored. A follow-up study is currently ongoing to demonstrate the antibody decay rates and the importance of booster vaccination. Furthermore, a study of cellular immunity, which plays a vital role in mediating host immune responses to vaccines, is underway. Lastly, since we enrolled only healthy participants, our findings may not be generalizable to individuals with substantial comorbidities or clinical frailty.

Conclusions
All five primary series of COVID-19 vaccine regimens were well-tolerated and had a favorable safety profile. All regimens elicited robust humoral immune responses against wild-type SARS-CoV-2 but attenuated responses against VoCs, particularly the Omicron strain, among healthy Thai populations. This study provides important evidence for planning an appropriate COVID-19 vaccination program, prioritizing vaccine allocation, and reducing vaccine hesitancy in low-income countries where vaccination coverage remains suboptimal and vaccine supplies are limited.
Supplementary Materials: The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/vaccines11030564/s1. Figure S1: Diagram of study participants; Figure S2: Anti-SARS-CoV-2 receptor-binding domain immunoglobulin G against wild-type SARS-CoV-2 following a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen and participant's age group; Figure S3: Neutralizing antibody against wild-type SARS-CoV-2 following a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen and participant's age group; Figure S4: Neutralizing antibody, based on the in-house surrogate virus neutralization test, against wild-type SARS-CoV-2 and other circulating variants of concern following the completion of a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen and participant's age group. Table S1: The proportion of seropositive participants based on anti-SARS-CoV-2 receptor-binding domain immunoglobulin G against wild-type SARS-CoV-2 following a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen; Table S2: The proportion of participants with positive neutralizing activity against wild-type SARS-CoV-2 following a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen; Table S3: The proportion of participants with positive neutralizing activity, based on the inhouse surrogate virus neutralization test, against wild-type and other circulating variants of concern of SARS-CoV-2 following the completion of a primary series of COVID-19 vaccination, stratified by the COVID-19 vaccine regimen.  Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data that were used and/or analyzed are available upon request from the corresponding author.