Immunogenicity and Safety of a 3-Dose Regimen of a SARS-CoV-2 Inactivated Vaccine in Adults: A Randomized, Double-Blind, Placebo-Controlled Phase 2 Trial

Abstract

The pandemic of coronavirus disease 2019 (COVID-19) caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has led to more than 250 million confirmed cases and more than 4.3 million deaths worldwide as of 13 August 2021 according to the World Health Organization (WHO) [1]. The significant morbidity and mortality call for urgent, effective vaccines and immunization strategies to contain this virus.

In the race to compete with virus spread, the vaccine industry has been collaborating with academy, health organizations, and regulatory agencies closely to accelerate vaccine development. Since the first-in-human clinical trial with a recombinant adenovirus type-5 vectored COVID-19 vaccine conducted in China in March 2020 [2], a variety of vaccines have been developed and undergone clinical assessment. As of 13 August 2021, there are 110 SARS-CoV-2 candidate vaccines in various phases of clinical development and another 184 in preclinical development, according to WHO [3]. Various platforms or technologies are applied for the development of COVID-19 vaccine, including inactivated vaccine, adenovirus vectored vaccine, recombinant protein-based vaccine, RNA vaccine, and DNA vaccine [3].

Given the urgent need for effective vaccines, 22 COVID-19 vaccines have received conditional approval or emergency use authorization (EUA) via accelerated licensure process so far [4]. In China, 7 COVID-19 vaccines have received conditional approval or EUA, including 5 inactivated vaccines, 1 adenovirus type-5 vectored vaccine, and 1 recombinant protein-based vaccine. These 5 authorized inactivated vaccines are manufactured using a similar process but differ in antigen content and virus seed. The mass vaccination programs with these vaccines have played a critical role for prevention and control of COVID-19.

The safety, immunogenicity, and/or efficacy in humans have been demonstrated and continued to be assessed in ongoing clinical trials for many of these vaccines [5–13]. However, studies have shown a significant trend of decline in antibody titer at various months following 2-dose vaccination with inactivated vaccine, adenovirus vectored vaccine, or mRNA vaccine [14, 15]. Some studies also demonstrated that a booster dose (third dose) elicited robust anamnestic response when given 6 months or more following the second dose [14].

Here we report the immunogenicity and safety of a 3-dose regimen of an inactivated SARS-CoV-2 vaccine (KCONVAC, manufactured by Shenzhen Kangtai Biological Products Co., Ltd., China, and Beijing Minhai Biotechnology Co., Ltd., China) from an ongoing phase 2 clinical trials conducted in Chinese adults and elders.

METHODS

Study Design and Participants

This 3-dose regimen study was a part of an ongoing phase 2, randomized, double-blind, and placebo-controlled clinical trial which was conducted in Jiangsu Provincial Center for Disease Control and Prevention (JPCDC). The study was done in accordance with the Declaration of Helsinki and Good Clinical Practice. An independent data safety monitoring board was established before the start of the trial to provide oversight of the safety data during the study. The protocol and informed consents were approved by the institutional review board of JPCDC. Written informed consent from all participants was obtained before screening for eligibility.

Eligible participants were healthy individuals aged 18 through 59 years, and 60 years or older who were seronegative for SARS-CoV-2 immunoglobulin M (IgM) and immunoglobulin G (IgG) and negative for SARS-CoV-2 nucleic acid as confirmed by pharyngeal swab reverse transcription polymerase chain reaction (RT-PCR). Confirmed cases, suspected cases, or asymptomatic cases of COVID-19 as referred to the Information System of China Disease Prevention and Control were excluded. Those who had close contact with confirmed or suspected cases, or had travel history to a foreign or domestic epidemic community within 14 days before vaccination were also excluded. To be included, participants should have an axillary temperature of 37.0°C or lower, and have general good health as established by medical history, physical examination, and laboratory testing. Pregnant or breastfeeding women were excluded. People with a previous SARS-CoV infection, mental disease, allergic reaction to any ingredient included in this vaccine or severe allergy to any other vaccines, congenital or acquired immune deficiency, human immunodeficiency virus infection, serious systemic diseases, or other major chronic illnesses were also excluded. A complete list of the inclusion and exclusion criteria is provided in the protocol. The protocol can be accessed by the following link: https://www.jscdc.cn/jkfw/kygz/202104/t20210425_70351.html.

Randomization and Masking

The vaccine strain of SARS-CoV-2 virus (19nCoV-CDC-Tan-Strain03, isolated in the Laboratory of National Institute for Viral Disease and Prevention, China Center for Disease Control, from clinical specimens obtained from SARS-CoV-2–positive patient) was cultivated in Vero cells. The harvested virus was inactivated by β-propiolactone, purified, and adsorbed to aluminum hydroxide (adjuvant). Each dose of vaccine contained 5 μg or 10 μg of total protein of inactivated SARS-CoV-2 virus and 0.25 mg of aluminum in a 0.5-mL liquid formulation. The placebo contained the same adjuvant but no viral protein. The experimental vaccines and placebo were blindly labelled with a randomization number on each vial as the only identifier.

The eligible participants were randomized within each age group (18 through 59 years, and 60 years or older) in a ratio of 2:2:1 to receive either 5 μg vaccine, 10 μg vaccine, or placebo. The randomization list was generated by an independent statistician using SAS software (version 9.4; SAS Institute). A unique randomization number in sequence was allocated to each participant, who then received a vaccine or placebo dose labelled with the same randomization number. The individuals involved in randomization and masking had no involvement in the rest of the trial. Participants, investigators, and staff undertaking laboratory testing were blinded to treatment allocation.

Procedures

Participants were administered 3 doses on day 0, 28 and 56, and observed for any immediate reaction for 30 minutes following each dosing. Diary cards were provided to participants to record any adverse events (AEs) occurred within 7 days after each dosing. Any AEs occurred from day 8 through day 28 after each dosing were also recorded. To verify the AEs, on-site visits were required on day 7 and 28 after each dosing. Telephone contacts were done by investigators on day 3 and 14 after each dosing. All AEs were graded according to the scale issued by the National Medical Products Administration, China in 2019 [16].

Blood samples for antibody assay were taken from all the participants before vaccination, and 28 days after the second and third doses. Binding antibody responses against the receptor binding domain (RBD-IgG) of the SARS-CoV-2 spike glycoprotein were tested by using enzyme-linked immunosorbent assay (ELISA) with a detection limit of 1:20. Serum samples were 2-fold serially diluted from 1:20 in 96-well plates coated with recombinant RBD, and incubated at 37°C for 60 minutes. Bound RBD-IgG was detected using a horseradish peroxidase-conjugated secondary antibody and substrate tetramethylbenzidine. Optical density was read at 450 nm and 630 nm to calculate the antibody titer. Live virus neutralizing antibody responses were measured by microneutralization assay with a detection limit of 1:4. In brief, heat-inactivated serum samples were serially diluted 2-fold from 1:4 in 96-well plates, and mixed with the same volume of live virus solution (strain 19nCoV-CDC-Tan-Strain03) of 100 CCID₅₀ (50% cell culture infectious dose), and then incubated at 37°C for 2 hours. Subsequently, Vero cell suspension containing 1.5 to 2.5 × 10⁴ cells was added to each well, and incubated at 37°C for 4 days. The neutralizing titer was defined as the reciprocal of the highest sample dilution that protected at least 50% of cells from cytopathic effect. Pseudovirus neutralization was tested using a vesicular stomatitis virus pseudovirus system expressing the spike glycoprotein as reported previously [17], with a detection limit of 1:10. Undetectable antibody titer was assigned a value of half the detection limit for calculation.

Outcomes

The primary end points for immunogenicity were neutralization antibody seroconversion and titer, and RBD-IgG seroconversion 28 days after the third dose. The secondary end points included the proportion of participants experiencing adverse reactions/events within 28 days following each vaccination, occurrence of serious AE (SAE) from the first dose through 12 months after the third dose, titer of RBD-IgG 28 days after the third dose, and the seroconversion and titer of RBD-IgG and neutralization antibody 28 days after the second dose.

Seroconversion was defined as antibody titer (1) < 1:4, < 1:30, or < 1:20 before vaccination and ≥1:4, ≥1:30, or ≥1:20 postvaccination; or (2) ≥1:4, ≥1:30, or ≥1:20 before vaccination and ≥4-fold higher postvaccination for neutralization antibody against live SARS-CoV-2, neutralization antibody against pseudovirus, or RBD-IgG, respectively.

Statistical Analysis

The sample size was determined based on the assumption that the seroconversion percentages for live neutralization antibody in vaccine group and placebo group were 80% and 30%, respectively. A sample size of 100 in the vaccine group versus 50 in the placebo group would have sufficient power (more than 99%) to demonstrate a real difference in the seroconversion percentages for live neutralization antibody between groups when tested with a 2-sided α value of .05.

All participants who received at least 1 dose were included in safety analysis. The number and proportion of participants experiencing adverse reactions or AEs in each group are presented. The immunogenicity analysis was done in per-protocol set consisting of participants who did not deviate from the eligibility criterion, received 2 doses, provided blood samples as scheduled, and had evaluable immunogenic data. Immunogenicity is expressed by seroconversion percentage, geometric mean titer (GMT), and the associated 95% confidence interval (CI). Antibody titers of individuals were log-transformed to calculate GMT per group. GMTs were compared with t test. The χ² test or Fisher exact test was used to compare difference between groups for categorical data. Analysis of variance was used to test the difference between groups for log-transformed antibody titers. All analyses were 2-tailed, and P < .05 was considered statistically significant. The study was registered with ChiCTR.org.cn, number ChiCTR2000039462.

RESULTS

The trial profile is shown in Figure 1. A total of 500 participants were enrolled, of whom 250 were 18–59 years of age and 250 were 60 years of age or older. In each age group, the eligible participants were randomized to receive 3 doses of 5 μg vaccine (n = 100), 10 μg vaccine (n = 100), or placebo (n = 50). All participants received at least 1 dose of vaccine or placebo and were included in safety analysis. In the age group of 18–59 years, 4 and 6 participants were not included in the population for immunogenicity analysis of the second and third doses, respectively, due to no immunogenicity data available. In the age group of 60 years or older, the corresponding number of participants were 17 and 24. Six participants aged 18–59 years and 24 participants aged 60 years or older discontinued the study. The average age across the 3 treatment groups were 38.8 to 42.8 years in age group 18–59 years and 64.6 to 64.9 years in age group 60 years or older. Baseline characteristics were generally similar between the 3 treatment groups in each age group (Table 1). This study is ongoing to continuously follow up the safety and antibody persistence as planned. The preliminary analysis reported here presents the data through the cutoff of 28 days after the third vaccination.

Figure 1.

Trial profile.

Open in new tabDownload slide

In the age group of 18–59 years, 23 (23%), 23 (23%), and 10 (20%) participants reported at least 1 AE, of whom 14 (14%), 16 (16%), and 6 (12%) participants reported at least 1 vaccination-related AE after receiving 5 μg vaccine, 10 μg vaccine, or placebo. Lower frequency of AE was reported in the age group of 60 years or older, where 4 (4%), 10 (10%), and 2 (4%) participants reported at least 1 AE, of whom 1 (1%), 10 (10%), and 2 (4%) participants reported at least 1 vaccination-related AE in the 3 treatment groups. All the AEs were grade 1 or 2 in intensity. No AE of grade 3 or higher was reported. The most common solicited injection-site AE across the treatment groups and age groups were pain. Solicited systemic AE was less frequent. A total of 5, 4, and 4 participants reported headache, fever, and fatigue, respectively (Table 2). Two SAEs (protrusion of intervertebral disc and inguinal hernia) were reported in 5 μg and 10 μg vaccine groups aged 60 years or older, respectively. Both were deemed by the investigators not related to study vaccination. No participant discontinued the study due to AE.

The baseline serostatus is summarized in Table 1. Before vaccination, all participants in the age group of 18–59 years were seronegative (under the detection limit) for live virus neutralizing antibody, pseudovirus neutralizing antibody, and RBD-IgG. Only 1 and 6 participants in the age group of 60 years or older were seropositive for pseudovirus neutralizing antibody and RBD-IgG, respectively. The vaccine induced significant antibody response (Table 3 and Figure 2). After vaccination with 2 doses of the investigational vaccine, the antibody titers were significantly elevated from the baseline with GMTs of 102 to 119, 170 to 176, and 1449 to 1617 for live virus neutralizing antibody, pseudovirus neutralizing antibody, and RBD-IgG, respectively, in the age group of 18–59 years. The GMTs for pseudovirus neutralizing antibody and RBD-IgG in the age group of 60 years or older were similar to those observed in the age group of 18–59 years, but GMTs for live virus neutralizing antibody in the age group of 60 years or older were approximately one-fourth of that in the age group of 18–59 years. Vaccination with the third dose of vaccine only slightly (less than 1.5-fold GMT) increased antibody titers across the treatment groups. At 28 days after the second dose of vaccine, 95% (92/97) to 100% (99/99) of participants aged 18 to 59 years, and 94% (86/92) to 100% (94/94) of participants aged 60 years or older underwent seroconversion for the 3 antibodies. Seroconversion percentages 28 days after the third dose of vaccines were generally similar to those after the second dose. As expected, placebo did not elicit obvious immune response in both age groups with only 0, 2, and 2 participants after the second dose, and 0, 1, and 0 participants after the third dose underwent seroconversion for the 3 antibodies. We also included live virus neutralizing antibody results from human convalescent serum (HCS) as a reference (Figure 2). The live virus neutralizing antibody titer induced by this investigational vaccine was higher in participants aged 18–59 years than that in HCS, but lower in participants aged 60 years or older than that in HCS.

Figure 2.

Antibody response after the second and third vaccination. A, Neutralizing antibody to live SARS-CoV-2. B, Neutralizing antibody to pseudovirus. C, RBD-IgG. Participants were vaccinated with the indicated dose levels of KCONVAC or placebo on day 0, 28, and 56. Blood samples were obtained 28 days after the second and third vaccination (ie, day 56 and 84). Horizontal bars show geometric mean titers, error bars indicate 95% confidence intervals, and dots indicate individual antibody titers. Abbreviations: HCS, human convalescent serum; RBD-IgG, immunoglobulin G binding antibody responses against the receptor binding domain; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Open in new tabDownload slide

DISCUSSION

With the intensive efforts exerted collaboratively by industry, academy, health organizations, and regulatory agencies, more and more vaccines against SARS-CoV-2 have been developed, undergone clinical assessment, and received conditional approval or EUA. Mass vaccination programs are being deployed around the world and are significantly contributing to the prevention of excessive morbidity and mortality. However, with emerging more transmissible and infectious variants of SARS-CoV-2 and time elapsing, the waning antibody titer might not be sufficient to prevent COVID-19 caused by the emerging variants, warranting a booster dose or third dose to provide adequate protection.

In this ongoing phase 2 trial, a 3-dose regimen of an inactivated vaccine was evaluated. The third dose of the 3-dose series cannot be deemed as a booster dose due to the short interval between the second and third dose, that is, it is a part of the primary series. This study on 3-dose regimen may provide us with a more insightful understanding of the antibody dynamics following vaccination with inactivated vaccines, and thus help develop an appropriate timing for vaccination with a third dose.

Administration of 3 doses of this investigational vaccine showed good safety profile at both 5 μg and 10 μg dosages in both younger and older adults, consistent with previous reports on 2 doses of inactivated SARS-CoV-2 vaccines [6, 7, 13]. The predominant AEs were injection-site pain. All the AEs were grade 1 or 2 in intensity. No AE of grade 3 or higher was reported. No SAE was deemed related to study vaccination. Lower frequency of AE was reported in older adults of 60 years or older, which is consistent with the experience from many other vaccines. As more safety data become available for various vaccines from clinical trials and postauthorization studies, it is more evident that inactivated SARS-CoV-2 vaccine is better tolerated in term of short-term safety profile as compared with mRNA vaccine or recombinant adenovirus-vectored vaccine [18–20].

Two doses of the investigational vaccine were able to induce significant antibody response for the 3 antibodies. Most participants in the both age groups underwent seroconversion for the 3 antibodies. The GMTs and seroconversion percentages observed in the participants aged 18 to 59 years reported here are similar to those reported previously in the same age group vaccinated with the same dosage of the investigational vaccine [13]. Vaccination with the third dose only slightly increase the antibody titers (less than 1.5-fold GMTs versus 28 days after the second dose/immediately before the third dose). The same 3-dose regimen (day 0, 28, and 56) with an inactivated vaccine was also reported recently, which showed a slight increase in neutralizing antibody titer (GMTs increased from 38.1 after the second dose to 49.7 after the third dose) [14]. It seems that a third dose with a relatively short dosing interval such as 28 days may not be helpful for induction of higher antibody titer. This is consistent with the interval between the first and second doses. Our previous report showed that a longer dosing interval between the first dose and the second dose (given at day 0 and 28 versus day 0 and 14) induced 1.6 to 3.5 times higher antibody titers 28 days after the second dose [13]. A pooled analysis of 4 randomized trials with ChAdOx1 nCoV-19 (AZD1222) vaccine also indicated that an interval of 12 or more weeks compared with an interval of less than 6 weeks between the first and second doses induced more than 2-fold higher binding antibody responses [21]. Previous experience with inactivated vaccines or recombinant protein vaccines further support that a longer dosing interval may be helpful to elicit higher antibody response, as was observed with human papillomavirus vaccine, inactivated polio vaccine, and inactivated hepatitis A vaccine [22–25]. All this evidence supports a longer interval between the second dose and the third dose for this inactivated vaccine to induce a robust anamnestic response.

Currently, accumulating data show that the antibody elicited by inactivated SARS-CoV-2 vaccine was not durable and persisted for 3 to 6 months [14, 26]. However, a booster dose (third dose) given 6 to 8 months later induced significantly elevated antibody titer, indicating that inactivated SARS-CoV-2 vaccine can elicit immune memory [14]. We are also studying the booster response to this investigational inactivated vaccine in a separate study cohort and will publish the results elsewhere.

When comparing the 2 age groups (18 to 59 years and 60 years or older), the investigational vaccine elicited similar antibody response in the 2 age groups with respect to pseudovirus neutralizing antibody and RBD-IgG, but GMTs for live virus neutralizing antibody were significantly lower in the age group of 60 years or older than in the age group of 18–59 years. This phenomenon was found after both the second and third dose.

Consistent with our previous finding [13], no significant dosage-dependent antibody response was observed. Both 5 μg and 10 μg doses elicited similar antibody response in both the age groups. This has supported the EUA for 5 μg dosage instead of 10 μg dosage.

Numerically, the GMT of RBD-IgG was higher than that of either live virus or pseudovirus neutralizing antibodies, and the GMT of pseudovirus neutralizing antibody was higher than that of live virus neutralizing antibody in both the age groups when the vaccines were administered using the same dosage and the antibody titers were assayed at the same time points, which is consistent with observations from previous reports [5, 13].

In the context that various platforms or technologies are applied by the currently available COVID-19 vaccines, it is interesting to compare the immune response induced by these vaccines. Although head-to-head comparison studies are not available, the accumulating reports indicate that inactivated vaccine is less immunogenic than mRNA vaccine or recombinant adenovirus-vectored vaccine [5–9, 11, 13, 27]. When comparing with human convalescent serum, this investigational vaccine induced higher antibody titer in participants aged 18–59 years but lower in participants aged 60 years or older. However, it should be noted that the human convalescent serum was not assayed concurrently with the blood samples from the study participants. Variation between lot-to-lot assays might exist.

With continued high global incidence, emerging more transmissible and infectious variants, and expected waning antibody titer with time, the vaccine industry and academia are increasingly interested in antibody persistence and the effect of a booster dose. Unfortunately, we have not yet completed our study on these topics, which are the objectives of our ongoing study and will be reported in the future.

In conclusion, the results reported here indicate that vaccination with 2 doses of this investigational inactivated vaccine induced robust immune response and had a good safety profile. A third dose of this inactivated vaccine given 28 days after the second dose had limited effect to further boost the antibody response. A third dose administered with a longer dosing interval may act as a booster dose, which requires further studies.

Notes

Acknowledgments. The authors appreciate the contributions of all investigators at the Jiangsu Provincial Center for Disease Control and Prevention (CDC) and Huaiyin CDC who worked on the trials. The authors thank Yuan-Zheng Qiu for helpful expert advice and support for writing. We thank all the participants who volunteered for this study.

Financial support. This work was supported by the Guangdong Emergency Program for Prevention and Control of COVID-19 (grant number 2020A1111340002); and the Shenzhen Key Research Project for Prevention and Control of COVID-19.

Potential conflicts of interest. J. L. is an employee of Shenzhen Kangtai Biological Products Co., Ltd. G. L., X. C., and Y. L. are employees of Beijing Minhai Biotechnology Co., Ltd. All other authors report no potential conflict of interests. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.

References

Author notes