Immunogenicity and safety of single booster dose of KD-414 inactivated COVID-19 vaccine in adults: An open-label, single-center, non-randomized, controlled study in Japan

ABSTRACT

Although vaccines for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) disease 2019 (COVID-19) induce effective immune responses, vaccination with booster doses is necessary because of waning immunity. We conducted an open-label, non-randomized, single-arm study in adults in Japan to assess the immunogenicity and safety of a single booster dose of the KD-414 purified whole-SARS-CoV-2-virion inactivated vaccine candidate after vaccination with a primary series of BNT162b2. The primary endpoint was serum neutralizing activity at 7 days after booster injection compared with the primary series of BNT162b2. The SARS-CoV-2-structural protein-binding antibody level and T cell response against SARS-CoV-2-Spike (S) peptides were also examined as secondary endpoints, and safety profile assessments were conducted. Twenty subjects who participated in a previous study declined an injection of KD-414 (non-KD-414 group) and received a booster dose of BNT162b2 instead. The non-KD-414 group was compared to the KD-414 group as a secondary outcome. A single dose of KD-414 induced lower serum neutralizing activity against the wild-type virus within 7 days compared to after the primary series of BNT162b2 but significantly induced anti-SARS-CoV-2-S1-receptor-binding domain-binding immunoglobulin G (IgG) antibodies and SARS-CoV-2-S peptide-specific CD4⁺ and CD8⁺ T cell responses. Local or systemic symptoms were significantly lower in the participants who received KD-414 than in those who received BNT162b2 as the third COVID-19 vaccine dose. The present data indicate that a single booster dose of KD-414 induces a substantial immune response in BNT162b2-primed individuals and has a good safety profile, thereby supporting further clinical trials to identify rational targets.

KEYWORDS:

• COVID-19
• SARS-CoV-2
• COVID-19 vaccine
• inactivated vaccine
• side effects
• adverse events
• vaccine safety
• neutralizing antibody
• KD-414

Introduction

The rapid spread of the novel coronavirus (severe acute respiratory syndrome coronavirus 2 [SARS-CoV-2]) disease 2019 (COVID-19) has resulted in the most serious global public health and socioeconomic disaster in this century.¹ To act against the pandemic, more than 160 COVID-19 vaccines are still under development² using multiple platforms.³ Among these, whole-virion inactivated vaccines directed toward Japanese encephalitis, hepatitis A, rabies, and poliovirus,⁴ represent one traditional approach. Unlike natural infection or live vaccines, inactivated vaccine-induced immunity is generally not long-lasting, and multiple doses are required to boost the effect over time.⁵ On the other hand, inactivated vaccines contain the intact pathogen. Thus, the immune response is likely to target viral structural proteins such as the matrix, envelope, and nucleoprotein,³ and to stimulate toll-like receptors and induce type 1 interferon (IFN).⁶

KD-414 is a purified inactivated whole SARS-CoV-2 virion-containing vaccine adjuvanted with aluminum hydroxide and developed by KM Biologics Co., Ltd. (Kumamoto, Japan).⁷ A double-blind, randomized, placebo-controlled phase I/II trial of KD-414 involving 210 healthy adults in Japan with no prior COVID-19 vaccination or history of COVID-19 showed good tolerability for all ages tested.⁷ A high dose (10 ug/dose) used in the phase 1/2 study to evaluate the efficacy of three doses induced significant neutralizing antibody activity in age-dependent manner. This was especially evident in participants under 40 years of age, when the first two doses were administered intramuscularly 28 days apart.⁷ Administration of the third dose of KD-414 after 6 or more months from the second dose was also well tolerated and induced a higher and longer-lasting neutralizing antibody response compared with after the second dose.⁷

More than 80% of Japanese people in their 20s and 30s have already completed the primary series of vaccinations, mostly using an mRNA platform.⁸ It is therefore important to evaluate the safety and immunogenicity of booster vaccinations. Here, we report the immunogenicity and safety of a single booster vaccination with KD-414 after primary vaccination with two doses of the BNT162b2 mRNA vaccine in Japanese volunteers. The aim was to determine whether a single booster dose of KD-414 after a primary series of BNT162b2 induces substantial SARS-CoV-2-specific humoral and cellular responses with a strong safety profile.

Materials and methods

Intervention; COVID-19 inactivated vaccine, KD-414

KD-414 is an inactivated vaccine containing purified inactivated whole SARS-CoV-2 virions cultured in Vero cells (RRID: CVCL_0059)⁹ that provides significant immunogenicity following two-dose primary series vaccination,⁷ especially in adults under 40 years of age. The vaccine is adjuvanted with aluminum hydroxide to boost the immune response. A SARS-CoV-2 strain isolated in Japan in 2020 was used for vaccine production.^7,9 One dose of KD-414 (0.5 mL) contains 10 μg of inactivated whole SARS-CoV-2 virions. The concentration of aluminum hydroxide in KD-414 is 0.40 mg/mL (as aluminum). The aluminum hydroxide used in KD-414 is the same as that used in the hepatitis B (HB) vaccine (Bimmugen) licensed in Japan.

Study design and participants

This open-label, non-randomized, single-arm study was conducted using 100 volunteer staff members of the National Center for Global Health and Medicine (NCGM) in Tokyo, Japan, from November 2021 to January 2022 (Certified Review Board of NCGM approval No. NCGM-C-004374, Japan Clinical Trial Registry No.: jRCTs031210388), as previously described.¹⁰

In brief, 100 individuals who participated in a clinical trial to evaluate the antigenicity and safety of the primary series of BNT162b2 (Ethics review committee of NCGM approval No. NCGM-A-004175)¹¹ and followed from March to June 2021 were selected as potential candidates (Figure 1(a)). Blood samples were collected at 7 (P_D7) and 40 (P_D40) days after the primary series of BNT162b2. This study intervention was performed 7.8 months after the second inoculation of BNT162b2 (Figure 1(a)). Blood samples were collected from individuals confirmed to be eligible and who agreed to participate in the present clinical study. All participants were monitored for participant assessed adverse reactions. Of these, those who requested KD-414 vaccination were vaccinated (KD-414 group) and those who did not request KD-414 vaccination had blood samples drawn and their antibody titers evaluated (non-KD-414 group). Subsequently, it was decided by the Japanese government that a third (single booster) dose of BNT162B2 would be administered to healthcare professionals in Japan, and thus all subjects in the KD-414 and non-KD-414 groups received a booster vaccination with BNT162b2 (Figure 1(a)).

Figure 1. Study protocol(a) and flow diagram(b).

Blood samples were obtained at 0 (immediately before the KD-414 shot, K_D0), 7 (K_D7), and 40 (K_D40) days after vaccination with the first dose of KD-414 (Figure 1(a)). Blood samples were subsequently obtained from the participants who had received a booster dose of BNT162b2 at 50 (7 days after the booster dose of BNT162b2, F_D7) and 83 (40 days after the booster dose of BNT162b2, F_D40) days after a single dose of KD-414 (Figure 1(a)). Whole blood specimens were subjected to T cell profiling analysis immediately after sample collection, and sera were stored at −80 °C until use.

All participants were also asked to record any solicited local (e.g., tenderness or spontaneous pain, erythema/redness, induration, swelling of the injection site) and constitutional symptoms (e.g., nausea/vomiting, headache, fatigue/malaise, myalgia, chills, diarrhea, arthralgia), body temperature, and antipyretic use every day for 7 days after each inoculation.

The kinetics of SARS-CoV-2-specific humoral and cellular responses and safety profiles were evaluated. This study was conducted in accordance with the Declaration of Helsinki¹² and the Clinical Research Act.¹³

Endpoints of the study

The primary endpoint of the study was the in vitro serum viral neutralizing titer at 7 days after vaccination compared with that after the primary series of BNT162b2. The secondary endpoints were (1) the serum neutralizing titer at 40 days after each vaccine injection, (2) the amount of anti-SARS-CoV-2-Spike (S) or -Nucleocapsid (NC)-binding antibodies, (3) CD4-positive (CD4⁺) and CD8-positive (CD8⁺) T cell responses against the SARS-CoV-2-S protein, (4) CD4⁺ T cell profiling, and (5) participant-assessed solicited symptoms and adverse effects for 7 days after each injection.

Statistical analysis

The main purpose of the present study was to assess the non-inferiority of serum-neutralizing activity after a single booster dose of KD-414 compared with that after the primary series of BNT162b2 at 7 days after the vaccination. Serum neutralizing activity after the primary series of BNT162b2 was used as a historical control. The lower bound (LB) of the 95% confidence interval (CI) for a geometric mean ratio (GMR) ratio>0.67 is often used as the non-inferiority limit for new COVID-19 vaccine candidates seeking approval based on the immunobridging approach in non-inferiority studies.¹⁴ As for the historical control, the population mean and standard deviation of the geometric mean of the 50% neutralizing titer (GM-NT50) was set to 4.25 and 0.4, respectively, on a common logarithm scale. Regarding 4.25, the geometric mean of the 50% neutralizing titer (GM-NT₅₀) on the common logarithm scale using the historical control as a population mean, and assuming a standard deviation (SD) 0.4 at K_D7 and a significance level of 2.5% for one side, 58 participants were required for 90% power and 44 for 80% power to assess with the LB of a 95% CI for a GMR>0.67. To assess the LB of a 95% CI for a GMR>0.67, 58 participants were required for 90% power and 44 for 80% power with a significant level of 2.5% for one side. Sixty of the participants (n = 100) enrolled in a previous study (NCGM approval No. NCGM-A-004175) were enrolled in the KD-414 group. The primary analysis population included members of the per-protocol population who received a booster KD-414 dose.

The proportion of T helper (T_h) cells and the results of interferon-gamma (IFN-γ) release assay (IGRA) and other obtained data were log-transformed before analysis. NT₅₀ values under the detection limit (≤40-fold dilution) were regarded as 20. GMR values were calculated based on the GM-NT₅₀ ratio between the groups or between time points, and the corresponding 95% CIs were determined. Immunogenicity data (e.g., serum neutralizing activity, SARS-CoV-2-S1-receptor-binding domain [RBD]-binding immunoglobulin G (IgG) level, IGRA results, and rate of T_h cell production) after KD-414 vaccination were compared with the values obtained from the non-KD-414 group and at different time points as the secondary endpoint analysis. The difference between groups or time points in the IGRA results was estimated by the difference in the medians, and the corresponding CIs were calculated using the bootstrap method. CIs for proportions of T_h cells were computed using the exact binomial (Clopper – Pearson) method. The correlation between neutralizing activity and the amount of SARS-CoV-2-S1-RBD-binding IgG was computed using Spearman’s correlation coefficient. Demographic characteristics were presented using the median and range for continuous variables or the number and proportion for categorical variables, and p values were calculated using Fisher’s exact test or Student’s t-test. Missing data were not imputed. A safety assessment was conducted for all participants who received a booster dose of KD-414. Descriptive summary data (numbers and percentages) were assessed for participants with any subjective adverse reaction up to 7 days after injection with KD-414. The incidence of self-reported solicited adverse events was compared with that after the third injection of BNT162b2 in the non-KD-414 group. The statistical significance threshold was p < 0.05. All analyses were performed using SASⓇ software, version 9.4 (Cary, NC, USA) and R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria).

Cells, viruses, and in vitro SARS-CoV-2 neutralizing assay

Transmembrane protease serine 2 (TMPRSS2)-overexpressing VeroE6 (VeroE6^TMPRSS2) cells (RRID: CVCL_YQ49) were obtained from the Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan). Two SARS-CoV-2 strains, wild-type (PANGO lineage B) and Delta (hCoV-19/Japan/TKYK01734/2021, PANGO lineage B.1.617.2, GISAID Accession ID: EPI_ISL_2080609), were isolated in March 2020 and April 2021, respectively, in Tokyo, Japan, as previously described.¹⁵ The SARS-CoV-2 neutralizing activity of sera samples was determined as previously described.^10,15,16 All samples used in the present study were evaluated in parallel. High neutralizing activity-confirmed convalescent plasma sample D43¹⁶ was used as a reference control to assess the intra-assay deviation.

Quantification of anti-SARS-CoV-2-S- and -NC-binding antibodies

Serum anti-SARS-CoV-2-S- and -NC-binding antibodies were examined as previously reported.¹⁰ In brief, the amount of anti-SARS-CoV-2-S1-RBD-binding IgG (IgG-S) was quantified using AdviceDx SARS-CoV-2 IgG II, while the presence of anti-SARS-CoV-2-S-binding immunoglobulin M (IgM-S) and anti-SARS-CoV-2-NC-binding IgG (IgG-N) was interpreted using AdviceDx SARS-CoV-2 IgM and SARS-CoV-2 IgG (Abbott, Chicago, IL), respectively, with the Architect apparatus (Abbott) based on a chemiluminescent microparticle immunoassay.

Peripheral T cell profiling using SARS-CoV-2-S peptides and flow cytometry analysis

Peripheral T cell responses against the SARS-CoV-2-S peptides were evaluated with the IGRA using a QuantiFERON SARS-CoV-2 RUO (QIAGEN, Hilden, Germany) according to the manufacturer’s instructions, as previously described.^10,17 An IFN-γ concentration of 0.20 IU/mL or higher (≥0.20 IU/mL) was interpreted as a positive T cell response, as previously reported.¹⁷

To characterize peripheral CD4⁺ T_h cells, phorbol 12-myristate 13-acetate and ionomycin (PMA-I)-stimulated CD4⁺ cells in the presence of brefeldin-A were stained with fluorescent-conjugated antibodies for human IFN-γ and interleukin-4 (IL-4), then subjected to flow cytometric analysis (SRL, Tokyo, Japan). T_h1, T_h2, and T_h0 (naïve T) cells were defined as IFN-γ⁺/IL-4^–, IFN-γ–/IL-4⁺, and IFN-γ⁺/IL-4⁺ CD4⁺ cells, respectively. The amount of each phenotype cell among all CD4⁺ cells and the ratio between T_h1 and T_h2 cells (T_h1/T_h2) were also calculated.

Results

Clinical characteristics of the participants

A CONSORT diagram of the present study is shown in Figure 1(b). Of the 100 candidates who received a primary series of BNT162b2 injections,¹⁰ 71 joined the clinical trial briefing and 69 freely and voluntarily agreed to participate in the present interventional study. All 69 of these participants were confirmed to meet the eligibility criteria; 49 (71.0%) agreed to receive a single dose of KD-414 (KD-414 group) and the remaining 20 (29.0%) did not (non-KD-414 group) (Figure 1(b)). Several individuals were excluded due to deviation from the study protocol. Forty-five individuals were injected with KD-414 and included in the final analysis as a per-protocol population, and 19 members of the non-KD-414 group were included in the comparison for reference. Most of the participants received an additional booster dose of BNT162b2 at 43 days after the KD-414 injection (Figure 1(a), S1). Blood samples, except for K_D0 for the non-KD-414 group, were collected intermittently for 83 days after a single dose of KD-414 (40 days after the booster dose of BNT162b2) (Figure 1(a)).

The demographic characteristics of the participants are shown in Table 1. No statistically significant differences in characteristics were found between the KD-414 and non-KD-414 groups, except for nine individuals in the KD-414 group who had certain medical histories such as hypertension (n = 3), asthma (n = 2), or atopic dermatitis (n = 1) (p = .0481) (Table 1).

Table 1. Characteristics of the participants.

	All (n = 64)	KD-414 group (n = 45)	No KD-414 group (n = 19)	p value
Age, years Median (range)	44.5 (23–74)	44 (24–64)	45 (23–74)	1.000*
Age group				.883*
20–29	8 (12.5%)	5 (11.1%)	3 (15.8%)
30–39	14 (21.9%)	11 (24.4%)	3 (15.8%)
40–49	20 (31.3%)	13 (28.9%)	7 (36.8%)
50–59	16 (20.5%)	12 (26.7%)	4 (21.1%)
≥60	6 (9.4%)	4 (8.9%)	2 (10.5%)
Sex				.965**
Male	22 (34.4%)	16 (35.6%)	6 (31.6%)
Female	42 (65.6%)	29 (64.4%)	13 (68.4%)
Type of job				.798*
Physician	4 (6.3%)	3 (6.7%)	1 (5.3%)
Nurse	8 (12.5%)	6 (13.3%)	2 (10.5%)
Allied health professional	1 (1.6%)	1 (2.2%)	0 (0.0%)
Administrative staff	11 (17.2%)	6 (13.3%)	5 (26.3%)
Other	40 (62.5%)	29 (64.4%)	11 (57.9%)
Underlying medical condition
Cardiovascular disease	0 (0.0%)	0 (0.0%)	0 (0.0%)	N/A
Kidney disease	0 (0.0%)	0 (0.0%)	0 (0.0%)	N/A
Liver disease	0 (0.0%)	0 (0.0%)	0 (0.0%)	N/A
Hematologic disease	1 (1.6%)	1 (2.2%)	0 (0.0%)	1.000*
Hemorrhagic disease	0 (0.0%)	0 (0.0%)	0 (0.0%)	N/A
Immunodeficiency disease	0 (0.0%)	0 (0.0%)	0 (0.0%)	N/A
Other	9 (14.1%)	9 (20.0%)	0 (0.0%)	.481*
Days to KD-414 vaccination Median (range)	203 (202–205)	203 (202–205)	N/A
Height, cm Median (range)	161.0 (150.0–180.0)	161.0 (153.0–180.0)	161.0 (150.0–180.0)	.451**
Weight, kg Median (range)	59.0 (38.0–98.0)	59.0 (42.0–76.0)	56.0 (38.0–98.0)	.333**
Body mass index, kg/m^2 Median (range)	21.8 (16.3–38.8)	21.7 (17.2–28.6)	22.0 (16.3–38.8)	.526**

*Fisher’s exact test, **Student’s t-test, N/A; not applicable.

A booster dose of KD-414 induced substantial serum neutralizing antibodies

Two or three doses of KD-414 as a primary series of vaccine injections demonstrate a favorable safety profile with substantial neutralizing antibody responses, especially in adults under 40 years of age.⁷ In the present study, we administered a single dose of KD-414 as a booster dose at 7.8 months after the primary series of BNT162b2 in 49 participants (Figure 1(a)). We first evaluated the kinetics of BNT162b2-induced serum neutralizing activity against wild-type SARS-CoV-2 (PANGO lineage B) during the observation period and found it gradually decreased after the primary series of vaccine injections (K_D0) (Figure 2(a)). In contrast, the single booster dose of KD-414 induced significant serum neutralizing activity against the wild-type virus within 7 days after the injection (Figure 2(a), S2(a)) but activity was significantly lower than that after the primary series of BNT162b2 (which did not exceed the non-inferiority GMR value) (Figure 2(b)). Activity was further increased 40 days after the KD-414 booster injection but continuously decreased in the non-KD-414 group, resulting in a much larger GMR value between groups (Figure 2(a), S2(b)). The elevated neutralizing activity at K_D40 did not exceed that at 40 days after the primary series of BNT162b2 (P_D40) but the GMR value was slightly lower than the non-inferiority GMR value (Figure 2(c)). The same trend was observed against the Delta variant (PANGO lineage B.1.617.2), which was the most dominant variant in Japan at the time of the intervention (Figure 2(d–f)). It is noteworthy that the serum neutralizing activity against the Delta variant was below the detection limit in all participants in the non-KD-414 group at K_D40, whereas KD-414 induced significant neutralizing activity against the variant (Figure S2(b), lower column).

Figure 2. Kinetics of serum SARS-CoV-2-neutralizing activity.

Red dots indicate the KD-414 group and blue dots indicate the non-KD-414 group. VeroE6^TMPRSS2 cells were exposed to (a–c) wild-type or (d–f) the SARS-CoV-2 Delta variant with or without various concentrations of diluted sera, and neutralizing activities were determined. The dashed line denotes the assay limit values (<40-fold). Black horizontal lines and vertical bars represent the geometric mean and standard deviation (SD) NT₅₀ values. A single booster dose of KD-414 induced significant neutralizing activity at 7 and 40 days after the vaccination against the wild-type (a) and Delta variant (d). (b, c, e, f) The numbers above the green dotted line represent the GMR values with 95% CIs in parentheses.

These results also confirmed, based on the kinetics of the IgG-S level (Figure S3(a)), the strong positive correlation observed between the NT₅₀ values against wild-type SARS-CoV-2 and the amount of IgG-S at all time points (Figure S3(e)). This suggests that the neutralizing activity after vaccination with KD-414 is largely due to the amount of IgG-S in the sera of booster-vaccinated participants.¹⁸ IgM-S was detected in many participants after the primary series of BNT162b2 (P_D7) (Supp. Table). On the other hand, a single dose of KD-414 had limited effects on IgM-S induction at 7 days after the booster dose of BNT162b2 (F_D7), with prevalence being significantly lower compared with after the primary series of BNT162b2. No IgG-N was detected throughout the study, including in the samples collected after a single dose of KD-414, suggesting that a single dose of KD-414 does not produce a detectable amount of IgG-N and that no study participants were infected with SARS-CoV-2 during the 40-day observation period (Supp Table).

A single booster dose of KD-414 as the third dose of COVID-19 vaccination enhanced the induction of neutralizing antibodies after injection of BNT162b2 as the fourth COVID-19 dose

The observation period was extended, and at 43 days after the single dose of KD-414, 47 participants (95.9%) in the KD-414 group and 14 (70%) in the non-KD-414 group received an additional booster dose of BNT162b2 (Figure 1(a)). The serum neutralizing activity at 7 (F_D7) and 40 (F_D40) days after the additional booster dose of BNT162b2, or 50 and 83 days after the single dose of KD-414, respectively, was evaluated. The additional single booster dose of BNT162b2 (fourth injection for the KD-414 group and third injection for the non-KD-414 group) induced drastically higher neutralizing activity compared with that after the primary series of BNT162b2 or the KD-414 booster injection (Figure 2(a)). It is noteworthy that the participants who received KD-414 showed higher neutralizing activity against both the wild-type and Delta variant (Figure S2(c,d)) than did the non-KD-414 group after the booster dose of BNT162b2, although no statistically significant differences were observed other than the activity against the Delta variant at F_D7 (Figure S2(c)).

KD-414 recalled SARS-CoV-2-S protein-specific CD4⁺ and CD8⁺ T cell responses

We next attempted to evaluate the kinetics of SARS-CoV-2-S peptide-specific CD4⁺ and/or CD8⁺ T cell responses across the observation period (Figure 3). In the 64 participants immunized with the primary series of BNT162b2, a substantial T cell response was observed within 7 days (at P_D7) in all the individuals with RBD-stimulated CD4⁺ (Ag1, Figure 3(a)) and S1/S2 subunit-stimulated CD4⁺ and CD8⁺ (Ag2, Figure 3(b)) T cell responses with positive IFN-γ production (≥0.20 IU/mL).^10,17 The responses gradually waned by 7.8 months after the primary series (Figure 3(a,b)). The single dose of KD-414 induced a substantial Ag1 response at KD7 and was further significantly elevated by KD40 (Figure 3(a)). The Ag2 response showed similar but relatively modest trends to that of Ag1 (Figure 3(b)). An additional booster dose of BNT162b2 re-stimulated both the Ag1 and Ag2 responses within 7 days after vaccination (F_D7), then decayed in a time-dependent manner within 40 days (F_D40), with no statistically significant differences seen between groups (Figure 3(a,b)).

Figure 3. CD4⁺ and/or CD8⁺ T cell responses after vaccination.

Red dots indicate the KD-414 group and blue dots indicate the non-KD-414 group. Heparinized whole blood samples were stimulated with major histocompatibility complex (MHC) class II SARS-CoV-2-S1-RBD peptides (a) or MHC class I SARS-CoV-2-S1/S2 subunit (including RBD) peptides (b), and the amount of released IFN-γ was determined. The IFN-γ release values (IU/mL) were log-transformed, and all negative values and zero were set to 0.01. Plasma IFN-γ production equal to or more than 0.20 IU/mL was considered a positive T cell response. (c) KD-414 vaccination did not affect the Th1/Th2 balance.

We also characterized peripheral CD4⁺ T_h cell responses stimulated with PMA-I on K_D7 and F_D7 using intracellular human IFN-γ and IL-4 flow cytometric analysis. No statistically significant differences in the percentage of T_h1 (IFN-γ⁺/IL-4^–) or T_h0 (IFN-γ⁺/IL-4⁺, naïve T) cells were found between the groups at K_D7 or F_D7 (Figure S4(a)). The population of T_h2 (IFN-γ^–/IL-4⁺) cells was slightly higher in the KD-414 group than in the non-KD-414 group at both K_D7 and F_D7 (Figure S4b), indicating a bias toward T_h2 cell responses at both K_D7 and F_D7 (Figure 3(c)); although, no statistically significant differences were observed between groups. The rate of naïve T cells production was comparable at both K_D7 and F_D7 (Figure S4(c)).

A booster dose of KD-414 had minimal systemic side effects compared with a booster dose of BNT162b2

Finally, we evaluated participant-assessed adverse effects and symptoms after each single booster dose of KD-414 and BNT162b2 for 40 days. The presence of local adverse events and subjective symptoms and whether body temperature exceeded 37.5 °C during the monitoring period after the third dose (KD-414 for the KD-414 group and BNT162b2 for the non-KD-414 group) were compared (Table 2).

Table 2. Self-reported adverse events.

	After KD-414 (KD-414 group) (n = 49)	After BNT162b2 (Non-KD-414 group) (n = 14)	p value*
Local injection site symptom
Pain	28 (57.1%)	14 (100%)	.0026
Erythema/redness	4 (8.2%)	3 (21.4%)	.1773
Induration	0 (0.0%)	2 (14.3%)	.0466
Swelling	0 (0.0%)	1 (7.1%)	.2222
Constitutional symptom
Body temperature ≥37.5 °C	0 (0.0%)	5 (35.7%)	.0003
Nausea/vomiting	4 (8.2%)	2 (14.3%)	.6071
Headache	10 (20.4%)	8 (57.1%)	.0158
Fatigue/malaise	11 (22.4%)	12 (85.7%)	<.0001
Myalgia	4 (8.2%)	8 (57.1%)	.0003
Chills	1 (2.0%)	4 (28.6%)	.0073
Diarrhea	1 (2.0%)	1 (7.1%)	.3978
Arthralgia	1 (2.0%)	5 (35.7%)	.0015
Any constitutional symptoms	19 (38.8%)	14 (100%)	<.0001
Antipyretic use			.0100
None	40 (81.6%)	6 (42.9%)
Routine or prophylactic use	2 (4.1%)	3 (21.4%)
Symptom management	7 (14.3%)	5 (35.7%)

*Fisher’s exact test.

For the 49 participants in the KD-414 group, pain at the injection site was the most common participant-assessed solicited local adverse event (Table 2), but the incidence was significantly lower than that after BNT162b2 in the non-KD-414 group (p = .0026). No induration or swelling at the injection site was reported after KD-414 administration (Table 2). None of the participants in the KD-414 group reported an elevated body temperature above 37.5 °C during the 7-day observation period after vaccination, in contrast with that after BNT162b2 administration (p = .0003) (Table 2). Further, the use of antipyretic agents (e.g., acetaminophen, nonsteroidal anti-inflammatory drugs [NSAIDs]), including routine or prophylactic use, was significantly lower in the KD-414 group (p = .0100) (Table 2). The incidence of systemic symptoms was significantly lower in the KD-414 group, with less than a half of the participants reporting at least one of the systemic symptoms described above after KD-414 injection (19 of 49, 38.8%), whereas all the individuals in the non-KD-414 group reported at least one of these symptoms after BNT162b2 (p < .0001) (Table 2).

From 8 through 40 days after KD-414 injection, three unspecified adverse events – urticaria, irregular menstruation, and elevated carcinoembryonic antigen (CEA)—were observed as adverse events with a causal or undeniable causal relationship to KD-414. The participants with elevated CEA levels had a medical history of lung cancer, and there was no suspicion of recurrence on periodic imaging diagnosis, leading the investigator to conclude that these elevated levels had an undeniable causal relationship to KD-414. All these adverse events were non-serious, and no serious adverse events were reported during the 40-day observation period after KD-414 injection.

Discussion

Increasing evidence suggests that booster vaccines do can be heterologous, especially mRNA-booster vaccines, because of their robust immunogenicity response elicited.^19–23 In contrast, boosting with adenovirus-vector vaccines (e.g., Ad26.COV2.S, AZD1222),^19,20,24 protein subunit vaccines (e.g., NVX-CoV2373),^25,26 or inactivated vaccine (e.g., CoronaVac)^27,28 results in relatively lower efficacy compared with mRNA-booster vaccines in individuals who have not received an mRNA vaccine as their primary series COVID-19 vaccine, especially with Omicron as the predominant circulating variant.^24,29

To our knowledge, no study has reported the immunogenicity or safety of heterologous immunization involving initial immunization with an mRNA vaccine followed by inactivated vaccine booster. We conducted the present open-label, non-randomized, single-arm study in Japan to evaluate the immune responses to and the safety of KD-414, an investigational new COVID-19 inactivated vaccine,⁷ after primary immunization with the BNT162b2 COVID-19 mRNA vaccine. We found that a single dose of KD-414 booster vaccine induced significant neutralizing antibodies against both the wild-type and Delta variant, especially at 40 days after the vaccination (Figure 2), eliciting a SARS-CoV-2-S-specific T cell response (Figure 3). When the KD-414-boosted participants received an additional booster dose of BNT162b2 as the fourth COVID-19 vaccine dose (two doses of BNT162b2 in the primary series, a single booster dose of KD-414, and a single booster dose of BNT162b2) at 43 days after KD-414 vaccination, the KD-414 group exhibited higher neutralizing activity against both the wild-type and Delta variant at 40 days after the booster dose of BNT162b2.

T cells play an important role in COVID-19 outcomes³⁰ and the maintenance of SARS-CoV-2 immunity, distinct from humoral immune kinetics.^31–33 Most T cell responses, whether induced by vaccination or natural infection, cross-recognize multiple amino acid mutations in variants, including the Omicron variant.³⁴ Therefore, the induction of T cell immunity against SARS-CoV-2 is an important goal for vaccine development,³⁵ especially in patients with B cell deficiencies due to, for example, hematopoietic malignancy or chemotherapy³⁶ and are thus more susceptible to a severe case of COVID-19. We observed the dynamic kinetics of the T cell response to SARS-CoV-2-S peptides during the study period using IGRA (Figure 3(a,b)). A significantly higher CD4⁺ T cell response to SARS-CoV-2-S1-RBD (Ag1) was induced at 40 days after KD-414 injection (K_D40) compared with the non-KD-414 group (Figure 3(b)). Importantly both BNT162b2 and KD-414 induced SARS-CoV-2-S1/S2 subunit-specific CD4⁺ and CD8⁺ responses (Figure 3(a,b)), and these T cell responses varied greatly between the participants. Individuals with a stronger response to the primary vaccination series tended to have a stronger T cell response after the booster dose, regardless of whether it was KD-414 or BNT162b2. The mechanisms underlying such high responses should be investigated further.

Previous studies of adverse effects have indicated more robust and broad immunity responses with certain heterologous vaccine combinations (e.g., AZD1222 priming followed by BNT162b2 or mRNA-1273 boosting), as higher systemic reactions may occur in such heterologous vaccine combinations.^27,37 Adenovirus-vector or inactivated vaccines are generally well tolerated.³⁸ In the present study, no local or systemic symptoms were reported more frequently by the participants who received a booster dose of KD-414 compared with those who received a booster dose of BNT162b2 (Table 2), which is a critical benefit for older adults, infants, and those who are reluctant for further vaccination because of adverse effects.³⁹

This study had some limitations. First, the sample size was smaller than initially planned (60 participants were planned, but only 49 participated). Second, this study was conducted as an open-label, non-randomized study; although the characteristics of the groups were mostly equivalent (Table 1). Third, since all the participants received an additional booster dose of BNT162b2 during the observation period, no long-term evaluation could be conducted. Fourth, we did not assess the viral neutralizing activity against multiple variants of concern, especially the Omicron variant, which has reduced susceptibility to neutralizing antibodies.^40–43 Further, no elevation of cytokines or chemokines such as IL-2, IL-4, IL-5, IL-10, IL-12, IL-13, granulocyte macrophage colony-stimulating factor (GM-CSF), IFN-γ, or tumor necrosis factor-α (TNF-α), was observed in the phase I/II clinical trial, even after administration of a primary series of KD-414;⁷ therefore, a further detailed analysis should be conducted to assess CD4⁺ and CD8⁺ T cell responses in booster dosing.

In conclusion, the safety and immunogenicity evaluation results obtained in the present study indicate that KD-414 is a promising inactivated vaccine candidate for heterologous immunization after mRNA vaccine priming. KM Biologics initiated several clinical trials at the end of April 2022: i) a phase III trial to compare the immunogenicity, efficacy, and safety of three doses of KD-414 as a primary series with that of the primary series of ChAdOx1 nCoV-19 (AZD1222) in individuals aged 18–40 years (jRCT2031210679); and ii) a phase II/III trial to determine the appropriate dosing of the primary series of KD-414 in children and teens aged 6 months to 17 years (jRCT2031220032). These studies should also provide further evidence of the efficacy, dosage, and rational target of the KD-414 vaccine for clinical application.

Contributor

Conceptualization; M.U., W.S., Data curation; N.T., Y.T., J.S.-T., Formal analysis; Y.S. and Y.U., Funding acquisition; M.U., K.S., W.S., Investigation; M.U., Y.T., J.S.-T., Ko.M., Ke.M., Methodology; Y.U., J.-T.H., A.M., S.Y., A.F., Project administration; J.T.-H., A.M., Supervision; H.M., N.O., Writing – original draft; Y.T., J.S.-T., Y.S., Y.U., J.T.-H., N.T.; Writing – review & editing; Y.T., J.T.-H., and W.S.

Acknowledgments

We thank all the individuals who participated in this study. This work was supported in part by KM Biologics Co., Ltd. and the Japan Agency for Medical Research and Development (AMED) (grant No. JP20fk0108417 to W.S.). The funders had no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. We thank Sho Saito, Katsue Watanabe, Yuriko Horikawa, and Shinsuke Kojima for managing the participants and administering the injections; Moto Kimura for collaboration with the company; Tetsuya Mizoue for providing information on previous studies; Akiko Kimura for coordinating the clinical research; Hiroko Aoshika and Keiko Hosokawa for assistance with the research administration; Yusuke Ohshiro for analyzing the specimens; Yasuharu Sasaki of the Japan Clinical Research Assist Center (JCRAC) Data Center for data management; Rina Yano for monitoring; Mariko Kato, Azusa Kamikawa, and Yumiko Kito for the technical assistance; and the Global Initiative for Infectious Diseases (GLIDE) support team for protocol development and administrative assistance.

Disclosure statement

Kengo Sonoda is an employee of KM Biologics Co., Ltd. All rights, titles, and interests to the patent of KD-414 have been assigned to KM Biologics Co., Ltd., Kumamoto, Japan. Mugen Ujiie received research funding from KM Biologics, Co., Ltd. for this and one other study. All other authors declare that they have no competing interests related to this study.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2023.2193074.

Additional information

Funding

The work was supported by the KM Biologics Kabushiki Kaisha .