The safety, immunogenicity, and efficacy of heterologous boosting with a SARS-CoV-2 mRNA vaccine (SYS6006) in Chinese participants aged 18 years or more: a randomized, open-label, active-controlled phase 3 trial

ABSTRACT

Continuous emergence of new variants of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), enhanced transmissibility, significant immune escape, and waning immunity call for booster vaccination. We evaluated the safety, immunogenicity, and efficacy of heterologous booster with a SARS-CoV-2 mRNA vaccine SYS6006 versus an active control vaccine in a randomized, open-label, active-controlled phase 3 trial in healthy adults aged 18 years or more who had received two or three doses of SARS-CoV-2 inactivated vaccine in China. The trial started in December 2022 and lasted for 6 months. The participants were randomized (overall ratio: 3:1) to receive one dose of SYS6006 (N = 2999) or an ancestral receptor binding region-based, alum-adjuvanted recombinant protein SARS-CoV-2 vaccine (N = 1000), including 520 participants in an immunogenicity subgroup. SYS6006 boosting showed good safety profiles with most AEs being grade 1 or 2, and induced robust wild-type and Omicron BA.5 neutralizing antibody response on Days 14 and 28, demonstrating immunogenicity superiority versus the control vaccine and meeting the primary objective. The relative vaccine efficacy against COVID-19 of any severity was 51.6% (95% CI, 35.5–63.7) for any variant, 66.8% (48.6–78.5) for BA.5, and 37.7% (2.4–60.3) for XBB, from Day 7 through Month 6. In the vaccinated and infected hybrid immune participants, the relative vaccine efficacy was 68.4% (31.1–85.5) against COVID-19 of any severity caused by a second infection. All COVID-19 cases were mild. SYS6006 heterologous boosting demonstrated good safety, superior immunogenicity and high efficacy against BA.5-associated COVID-19, and protected against XBB-associated COVID-19, particularly in the hybrid immune population.

Trial registration: Chinese Clinical Trial Registry: ChiCTR2200066941

KEYWORDS:

• SARS-CoV-2
• mRNA vaccine
• heterologous boosting
• safety
• immunogenicity
• efficacy

Introduction

Since December 2019, the coronavirus disease 2019 (COVID-19) pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has emerged as a major challenge for global public health. Vaccination is critical in reducing COVID-19-associated morbidity and mortality [1]. However, the continuous emergence of SARS-CoV-2 new variants caused enhanced transmissibility and/or immune escape, particularly after Omicron variants emerged in November 2021 [2,3]. Along with the waning immunity after vaccination, these new variants weakened the protection conferred by the primary vaccination. Boosting is needed [4,5].

Studies showed that heterologous boosting was safe and more immunogenic than homologous boosting [6–9], and demonstrated effectiveness in real-world [10–12]. However, efficacy data on heterologous boosting are rare, particularly those on preventing Omicron BA.5 and XBB are unavailable.

After China changed its “zero COVID-19” policy in December 2022, there were two outbreaks caused by BA.5 and XBB in China from December 2022 to June 2023 [13]. We evaluated the safety, immunogenicity, and efficacy of heterologous boosting with a SARS-CoV-2 mRNA vaccine (SYS6006) in Chinese healthy adults aged 18 years or more across the two outbreaks, using a licensed recombinant protein vaccine as the control.

Materials and methods

Study design and participants

This randomized, open-label, active-controlled trial was performed at The First Hospital of Hebei Medical University, Shijiazhuang, China, on 11 December 2022. The protocol and informed consent form were reviewed and approved by the Research Ethics Committee of this hospital. Written informed consent was obtained from each participant before screening. This trial was done per the principles of the Declaration of Helsinki and Good Clinical Practice.

Eligible participants were healthy adults aged 18 years or more who had received two or three doses of SARS-CoV-2 inactivated vaccine at least 6 months before the enrolment, negative for SARS-CoV-2 antigen test on the day of enrolment. To be eligible for the immunogenicity subgroup, the participants should also be negative for the SARS-CoV-2 reverse transcription polymerase chain reaction (RT–PCR) test within 48 h before enrolment. Key exclusion criteria included: history of SARS-CoV-2 infection within 6 months, history of hypersensitivity to any component of the investigational vaccine or history of severe allergic reactions to vaccines or drugs, axillary temperature ≥37.3°C on the day of enrolment or within 24 h before enrolment, active malignant tumours, severe or uncontrollable medical conditions, history of treatment with immunosuppressive or immunomodulatory agents, blood products or immunoglobulins. Pregnant or breastfeeding women were excluded. More details are provided in the supplementary material.

Randomization

The first 520 participants were enrolled into the immunogenicity subgroup and randomized (1:1), and the rest participants were subsequently randomized (∼3.7:1), to give an overall randomization ratio of 3:1 between the SYS6006 and control groups. The ratio of ∼3.7:1 was achieved by the dynamic blocked randomization method (110 blocks of 3:1 with a block size of 8, and 260 blocks of 4:1 with a block size of 10), as pre-specified in the randomization plan. The randomization list was generated by an independent statistician using SAS 9.4. Although the study was designed on an open label, the participants blindly received random numbers. After randomization and immediately before vaccination, the treatment assignment could be revealed.

Trial procedures

SYS6006 (CSPC Megalith Biopharmaceutical Co. Ltd, Shijiazhuang, China) was a lipid nanoparticle (LNP)-coated SARS-CoV-2 mRNA vaccine containing 30-μg mRNA/0.3 mL/dose which was designed based on the ancestral spike (S) protein, and harboured several key mutations presented in Delta, Omicron BA.4/5, and BF.7, as described previously [14]. This vaccine was validated in human cells in vitro, mice and non-human primates in preclinical studies [14], and assessed in previous clinical trials [15]. The sequence and mutations introduced to SYS6006 were reported previously [16]. The LNP consisted of ionizable lipids, 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC), cholesterol, and mPEG-DMG-2 K [17]. The control vaccine ZF2001 (Zhifei Longcom Biopharmaceutical Co., Ltd, Hefei, China) was a SARS-CoV-2 recombinant protein vaccine containing 25-μg ancestral (HCoV/Wuhan/IVDC-HB-01) receptor-binding domain (RBD) (residues 319–537, accession number YP_009724390) dimer and 0.25-mg aluminium hydroxide in 0.5 mL/dose, as described previously [18]. Participants received one heterologous boosting dose by intramuscular injection.

Participants were observed for 30 min following vaccination for immediate reaction. Diary cards were used to record solicited injection-site and systemic adverse events (AEs) for 14 days and unsolicited AEs for 28 days.

The immunogenicity subgroup included 520 participants. Blood samples were taken on Days 0, 14, 28, 90 and 180. Neutralizing antibodies (Nabs) were measured by cytopathic effect-based live virus microneutralization assay against wild type (WT) and BA.5. WT and BA.5 Binding antibodies (IgG) were measured by enzyme-linked immunosorbent assay (ELISA) (Supplementary Material).

COVID-19 case surveillance was driven by suspected symptoms according to the Food and Drug Administration (FDA) [19]. If any suspected symptom appeared, a self-tested antigen test was done to distinguish infection-associated symptoms from vaccination-associated systemic AEs. If positive, nasal/oropharyngeal swabs were collected for reverse transcription-polymerase chain reaction (RT–PCR) test. The swabs positive for RT–PCR test would be sequenced and typed (Supplementary Material).

Endpoints

The safety endpoints included solicited injection-site and systemic AEs within 14 days and unsolicited AEs within 28 days after vaccination, serious AEs (SAEs) and AEs of special interest (AESIs) during the study. AE grading was done per the guidelines of the Chinese National Medical Products Administration (Table S1) [20].

The immunogenicity endpoints included serum geometric mean titres (GMTs), seroconversion rates (SCRs), and geometric mean fold increase (GMFI) of Nab and IgG against WT and BA.5 on Days 14, 28, 90, and 180.

The efficacy endpoint was RT–PCR-confirmed COVID-19 case of any severity or severity and above with onset at least 7 days or 14 days after vaccination. The severity was graded per the FDA’s criteria [19]. The case definition was the presence of at least one suspected COVID-19 symptom, and a positive RT–PCR test in a nasal/oropharyngeal specimen obtained during the symptomatic period or within 4 days before or after it. Post hoc efficacy endpoints included RT–PCR-confirmed COVID-19 cases caused by a second infection, and antigen-positive COVID-19 cases.

Statistical analysis

The sample size for safety was not based on formal statistical hypothesis testing. A sample size of 3000 will have a 95.0% probability of observing at least one AE with an incidence of 0.1%. The sample size for immunogenicity was to test the hypothesis that the Nab GMT induced by SYS6006 was superior to that induced by ZF2001 on Day 14. The assumed GMT ratio of SYS6006 to ZF2001 was 1.5:1. The log-transformed standard deviation of Nab was 0.6. With an allocation ratio of 1:1, a superiority margin of 1.0, and non-evaluable participants of 5%, a sample size of 520 could provide more than 90% power at a one-sided significance level of 0.025. In the evolving pandemic context when the trial was initiated, it was hard to make a reliable estimation of COVID-19 incidence for efficacy assessment. Therefore, we did not estimate the power of efficacy assessment when developing the protocol. The sample size of 4000 was for preliminary efficacy assessment. However, with the surge of COVID-19 incidence in China since December 2022, the sample size would provide sufficient power for efficacy assessment.

The safety analysis included all vaccinated participants. The primary immunogenicity analysis was done in the per-protocol set (PPS) who did not deviate from the eligibility criteria were randomized and vaccinated per the protocol, had no conditions interfering with immune response, and had evaluable pre-vaccination and at least one post-vaccination antibody titres. Immunogenicity superiority of SYS6006 to ZF2001 was demonstrated if the lower bound of 95%CI for the adjusted GMT ratio was >1.0.

The primary efficacy analysis was done in the PPS population who were RT–PCR negative at baseline received a vaccination, had at least one follow-up for COVID-19 case 7 or 14 days after vaccination, and had no conditions interfering with efficacy assessment. The full analysis set (FAS) included all randomized and vaccinated participants per the intention-to-treat principle. Vaccine efficacy (VE) relative to the active control was estimated by (1-incidence rates ratio) × 100%.

Nab titres were log-transformed to calculate GMT and GMFI. Adjusted GMT was calculated by the analysis of covariance (ANCOVA), where treatment allocation was a fixed effect, and baseline GMT was covariate. The statistical significance was tested by t-test or analysis of variance. The SCR difference between groups was tested by the chi-square test or the Fisher exact probability method. SAS 9.4 was used. All analyses were two-tailed. P < 0.05 was considered statistically significant.

Results

Participants

From 11–20 December 2022, 4492 volunteers were screened, and 3999 were enrolled and randomized to SYS6006 (N = 2999) or ZF2001 (N = 1000) groups. The first 520 participants were included in the immunogenicity subgroup with 260 in each group. A total of 3,987 participants received vaccination (SYS6006: 2,994; ZF2001: 993). Twelve participants withdrew informed consent after randomization (Figure 1).

Figure 1. Trial profile. Two additional participants included in the SYS6006 group were from the control vaccine group due to vaccination error (*).

The baseline demography was generally similar between the two groups regarding age, sex, body mass index, and medical and vaccination history. All the participants were Chinese in ethnicity. The average duration since the last SARS-CoV-2 vaccination was 369.0 or 367.0 days. 83.7% or 85.1% of the participants received three SARS-CoV-2 vaccinations (Table 1).

Table 1. Baseline characteristics of the study participants.

	All randomized participants		Immunogenicity subset
	SYS6006 (N = 2999)	Control (N = 1000)	SYS6006 (N = 260)	Control (N = 260)
Age, years
Mean (SD)	37.0 (12.3)	36.7 (12.4)	39.0 (11.7)	37.9 (12.5)
Median	35.0	35.0	36.0	35.0
n (Missing)	2997 (2)	998 (2)	260 (0)	260 (0)
Age group – no. (%)
18–59y	2870 (95.8)	950 (95.3)	243(93.5)	243(93.5)
≥ 60y	127 (4.2)	47 (4.7)	17(6.5)	17(6.5)
n (Missing)*	2997 (2)	997 (2)	260 (0)	260 (0)
Sex – no. (%)
Male	1950 (65.1)	635 (63.6)	133 (51.2)	139 (53.5)
Female	1047 (34.9)	363 (36.4)	127 (48.8)	121 (46.5)
n (Missing)	2997 (2)	998 (2)	260 (0)	260 (0)
Body-mass index, kg/m^2
Mean (SD)	25.4 (4.4)	25.3 (4.2)	25.1 (3.7)	25.0 (4.0)
Median	25.0	24.8	24.6	24.3
Range	15.2, 48.2	15.9, 42.3	16.3, 40.9	17.5, 40.6
n (Missing)	2994 (5)	995 (5)	260 (0)	260 (0)
Medical history – no. (%)
Yes	439 (14.6)	153 (15.3)	64 (24.6)	54 (20.8)
No	2560 (85.4)	847 (84.7)	196 (75.4)	206 (79.2)
Days since the last dose of SARS-CoV-2 vaccine
Mean (SD)	369.0 (82.1)	367.0 (80.1)	359.2 (74.5)	358.7 (68.1)
Median	363.0	364.0	358.0	363.0
Range	49, 657	184, 620	195, 584	184, 575
n (Missing)	2992 (7)	994 (6)	260 (0)	259 (1)
SARS-CoV-2 vaccination history – no. (%)
2 doses	484 (16.1)	146 (14.6)	24 (9.2)	22 (8.5)
3 doses	2509 (83.7)	851 (85.1)	235 (90.4)	237 (91.2)
Others^#	6 (0.2)	3 (0.3)	1 (0.4)	1 (0.4)

*One participant, who was 16 years old, was excluded from the calculation.

^# Five participants had an unvaccinated history of the COVID-19 vaccine and one participant had a SARS-CoV-2 vaccination history of 4 doses in the SYS6006 group, while three participants had an unvaccinated history of COVID-19 vaccine in the control group. These participants were excluded from per-protocol analysis.

Safety

The safety analysis included 3987 vaccinated participants. The raw incidences of AEs were 67.3% and 46.5% within 28 days following SYS6006 or ZF2001 vaccination, respectively. Most AEs were grade 1 or 2. Three participants reported 3 unrelated grade-4 AEs (SYS6006: fever, cerebral infarction; ZF2001: fever). The two cases of grade-4 fever had a maximum temperature of more than 39.5°C for three consecutive days, which were deemed related to SARS-CoV-2 infection by the investigators. The case of cerebral infarction had a history of cerebral infarction, long-term history of hypertension and hyperlipidaemia. No grade-5 AE (death) was reported. Throughout the study, 22 (0.7%) and 6 (0.6%) participants reported SAEs in the SYS6006 and ZF2001 groups, respectively. All were unrelated to vaccination. No AESI or vaccine-associated enhanced disease was observed.

By excluding the participants (837 or 21.0%), who were infected with SARS-CoV-2 within 4 days of AE symptomatic period, 59.6% (1445/2424) and 27.0% (196/726) of participants reported AE within 28 days following SYS6006 or ZF2001 vaccination, respectively. The solicited systemic and injection-site AE incidences were 38.0% and 48.6% in the SYS6006 group, and 17.2% and 14.9% in the ZF2001 group. The most common solicited systemic and injection-site AEs were fever (SYS6006 group: 29.4%; ZF2001 group: 9.5%) and pain (SYS6006 group: 45.8%; ZF2001 group: 14.2%), respectively. The incidences of grade-3 fever were 7.2% (SYS6006) and 3.0% (ZF2001) (Figure 2).

Figure 2. Adverse Event Following Booster Vaccination. The number above the bar indicates the proportions of participants who reported the indicated solicited AEs (D0-D14) and unsolicited AEs (D0-D28) with intensity in all vaccinated participants who received SYS6006 (n = 2994) or the control vaccine (n = 993) (A), in the vaccinated participants (SYS6006: n = 2424; the control vaccine: n=726) who were not infected with SARS-CoV-2 within 4 days of AE symptomatic period (B), and in the vaccinated participants (SYS6006: n = 2080; the control vaccine: n = 603) who were not infected with SARS-CoV-2 (C).

By excluding the participants (1304 or 32.7%), who were infected with SARS-CoV-2 during the study, 59.6% (1240/2080) and 26.0% (157/603) of participants reported AE within 28 days following SYS6006 or ZF2001 vaccination, respectively. The solicited systemic and injection-site AE incidences were 38.2% and 47.8% in the SYS6006 group, and 17.4% and 13.1% in the ZF2001 group. The safety profile was similar to that excluding the participants who were infected with SARS-CoV-2 within 4 days of the AE symptomatic period (Figure 2).

Immunogenicity

The PPS immunogenicity analysis included 262 participants (SYS6006: 139; ZF2001:123). The predominant reasons for exclusion from this analysis were RT–PCR or antigen-positive before blood collection (Table S2). Before vaccination, most participants (SYS6006: 92.1%; ZF2001: 95.1%) were seropositive for WT Nabs, and fewer participants (SYS6006: 15.1%; ZF2001: 13.8%) were seropositive for BA.5 Nabs. SYS6006 boosting induced significant WT Nabs titres which increased from 22.7 (baseline) to 1174.6 (Day 14) and 1252.4 (Day 28), corresponding to 51.7- and 54.6-fold increase, respectively. BA.5 Nabs titres also significantly increased from 2.8 (baseline) to 235.6 (Day 14) and 155.5 (Day 28), corresponding to 83.1- and 54.6-fold increase, respectively. WT and BA.5 Nabs persisted for at least six months (Figure 3).

Figure 3. Live virus neutralizing antibody responses following booster vaccination. Live virus neutralizing antibody against wild type (A) or Omicron BA.5 (B). Interleaved scatter plot: geometric mean titre with 95% CI. The numbers above the scatter plots are the geometric mean titres. The numbers in the parentheses are the geometric mean fold increases versus baseline. The p-values indicate the statistical significance for geometric mean titres between SYS6006 and the control vaccine. The horizontal dot lines are the lower limit of quantification of the specific assay.

Considering different baseline GMT between groups, we calculated ANCOVA-adjusted GMT ratios of SYS6006 to ZF2001 for WT Nabs which were 4.0 (95% CI, 2.9-5.6) on Day 14 and 2.7 (95% CI, 2.0-3.7) on Day 28. The immunogenicity superiority of SYS6006 versus ZF2001 was demonstrated per the pre-defined criterion. The ANCOVA-adjusted GMT ratios for BA.5 Nabs were 4.4 (95% CI, 3.0–6.5) on Day 14 and 2.7 (95% CI, 1.8–4.0) on Day 28, also demonstrating immunogenicity superiority (Table S3). SYS6006 boosting induced significantly higher IgG responses than ZF2001 did, except those on Day 180 (Figure S1).

Efficacy

COVID-19 case surveillance lasted for 6 months. Per-protocol analysis started from Days 7 or 14 through the end of the trial. The predominant reason for exclusion from PPS analysis was positive RT–PCR or antigen test before case surveillance began (Table S4).

Of the 3987 vaccinated participants, 2130 (53%) experienced suspected symptoms, of whom 1712 had symptoms onset during Day 0–6 and 418 during Day 7–180. Per the protocol, symptoms onset during Day 0–6 did not trigger RT–PCR nasal/oropharyngeal sampling. Of the 418 participants, 254 (61%) underwent RT–PCR sampling (SYS6006: 140 or 64%; ZF2001: 114 or 57%), and 164 (39%) were not RT–PCR sampled (Figure S2).

From Day 7 to Month 6, 198 (SYS6006: 123; ZF2001: 75) RT–PCR-confirmed COVID-19 cases of any severity were detected in the PPS, yielding a relative VE of 51.6% (95% CI, 35.5–63.7) against any variant, 66.8% (95% CI, 48.6-78.5) against BA.5, and 37.7% (95% CI, 2.4–60.3) against XBB. From Day 14 to Month 6, 122 (SYS6006: 81; ZF2001: 41) RT–PCR-confirmed COVID-19 cases of any severity were detected in the PPS, yielding a relative VE of 42.1% (95% CI, 15.7–60.2) against any variant, 84.0% (95% CI, 56.8-94.1) against BA.5, and 30.4% (95% CI, −12.2–56.8) against XBB. During the study, 239 (SYS6006: 145; ZF2001: 94) RT–PCR-confirmed COVID-19 cases of any severity were detected in the FAS, yielding a relative VE of 54.2% (95% CI, 40.7–64.7) against any variant, 64.8% (95% CI, 47.5–76.4) against BA.5, and 46.6% (95% CI, 19.7–64.4) against XBB (Figure 4). The proportions of SARS-CoV-2 variants detected in the RT–PCR-confirmed COVID-19 cases are shown in Figure S3.

Figure 4. Kaplan-Meier plots of the efficacy of SYS6006 against RT-PCR-confirmed COVID-19 of any severity. Cumulative incidence of RT-PCR-confirmed COVID-19 of any severity with onset at least 7 days after vaccination in the per-protocol population (A), with onset at least 14 days after vaccination in the per-protocol population (B), and with onset after vaccination in the full analysis set population (C). Relative vaccine efficacy was estimated by (1-IRR) × 100%, where IRR is the incidence rates ratio (SYS6006 to the control vaccine) of RT-PCR-confirmed COVID-19 cases per 1000 person-years. *Incidence Rate was calculated per 1000 person-years with no. of cases divided by the surveillance time. The surveillance time was the total time of person-years for the given endpoint across all participants within each group at risk for the endpoint. The period for COVID-19 case accrual was from the beginning of case follow-up through Month 6 (the end of this study). ^#Sub-lineages of BA.5.2.48 including DY.2; Sub-lineages of BF.7.14 including BF.7.14.1, BF.7.14.4 and BF.7.14.5; Sub-lineages of XBB.1.5 including GF.1; Sub-lineages of XBB.1.9 including EG.2, EG.5.1, FL.2.3, FL.3, FL.4, FL.5; Sub-lineages of XBB.1.16 including FU.1; ^§Others of XBB including XBB.1.22, XBB.1.17.1, XBB.1.24.3, XBB.1.43.1, XBB.2.3.6, FY.3.1, and FP.4; ^&Others that are not included in BA.5 or XBB including BN.1.3.5, FR.1.1, FR.1.3.

Post hoc analysis showed that 26 (SYS6006: 11; ZF2001: 15) RT–PCR-confirmed COVID-19 cases of any severity caused by a second infection were detected during the study, yielding a relative VE of 68.4% (95% CI, 31.1–85.5). All were infected by XBB. From Days 0, 7 or 14 to the end of study, 602, 187 or 123 antigen-positive COVID-19 cases of any severity were accrued in the SYS6006 group, and 313, 111 or 67 cases were accrued in the ZF2001 group, yielding a relative VE of 43.8% (95% CI, 35.6–51.0), 50.5% (95% CI, 37.4–60.9) or 46.2% (95% CI, 27.5–60.0), respectively (Table 2).

Table 2. Post hoc analysis for vaccine efficacy.

Efficacy Endpoint	SYS6006		Control		Relative Vaccine Efficacy, % (95% CI)
	No. of Cases	Incidence Rate^# (95% CI)	No. of Cases	Incidence Rate^# (95% CI)
RT-PCR Confirmed COVID-19 caused by a second infection	n = 889		n = 381
XBB	11	31.5 (17.4, 56.9)	15	99.5 (60.0, 165.0)	68.4 (31.1, 85.5)
XBB.1.5 & Sub-lineages	1	2.9 (0.4, 20.3)	1	6.6 (0.9, 46.8)	56.7 (−592.6, 97.3)
XBB.1.9 & Sub-lineages	3	8.6 (2.8, 26.6)	4	26.4 (9.9, 70.4)	67.5 (−45.1, 92.7)
XBB.1.16 & Sub-lineages	3	8.6 (2.8, 26.6)	9	59.5 (31.0, 114.4)	85.6 (46.8, 96.1)
XBB.1.31	1	2.9 (0.4, 20.3)	1	6.6 (0.9, 46.8)	56.7 (−592.7, 97.3)
Others	3	8.6 (1.8, 25.1)	0	0.0 (0.0, 24.3)	NE
Antigen-positive COVID-19
Day 0 to Month 6 (FAS)	n = 2992		n = 995
	602	540.8 (499.3, 585.7)	313	962.7 (861.8, 1075.5)	43.8 (35.6, 51.0)
Day 7 to Month 6 (PPS)	n = 2235		n = 699
	187	204.7 (177.4, 236.3)	111	413.8 (343.6, 498.4)	50.5 (37.4, 60.9)
Day 14 to Month 6 (PPS)	n = 2161		n = 653
	123	141.1 (118.3, 168.4)	67	262.2 (206.4, 333.2)	46.2 (27.5, 60.0)

#Incidence Rate was calculated per 1000 person-years with no. of cases divided by the surveillance time. The surveillance time was the total time of person-years for the given endpoint across all participants within each group at risk for the endpoint. The period for COVID-19 case accrual was from the beginning of case follow-up through Month 6 (the end of this study).

RT-PCR-confirmed COVID-19 cases caused by a second infection were defined as participants who were vaccinated and got the first infection (antigen- or RT-PCR positive) before COVID-19 case accrual and were RT-PCR-confirmed symptomatic COVID-19 case upon the second infection. The case accrual began from the date for the first infection (for baseline RT-PCR negative) or the vaccination date (for baseline RT-PCR positive). Other exclusion criteria for this analysis are the same as those for PPS analysis.

Antigen-positive COVID-19 cases were defined as the occurrence of at least one suspected COVID-19 symptom occurring after vaccination and a positive antigen test obtained during the symptomatic period or within 4 days before or after it.

All the COVID-19 cases were mild. No severe case was observed during the study.

Discussion

The COVID-19 pandemic constitutes a huge challenge for global health. Continuous emergence of new variants enhanced transmissibility, significant immune escape, and waning immunity call for booster vaccination.

This study showed that SYS6006 heterologous boosting had an acceptable safety profile. We did not include an mRNA comparator; therefore; the AE incidence could not be compared directly with other mRNA vaccines. The AE incidence following SYS6006 boosting was within the range for the widely used or newly approved mRNA vaccines boosting (local AE incidence: 76.0%−97.0%; systemic AE incidence: 30.0%−79.7%) [6,21,22]. Because most COVID-19 symptoms were overlapping with the systemic AE following vaccination, coincident COVID-19 outbreak after vaccination affected the AE profile, specifically augmented the incidence and severity of systemic AE. Both groups had high fever incidence and high proportions of grade-3 fever in all vaccinated participants. However, by excluding the participants who were infected with SARS-CoV-2, fever incidence and proportions of grade-3 fever declined obviously, particularly for ZF2001. Therefore, it would be more accurate to assess vaccine safety in this population. A similar trend was seen for other systemic AEs. Because the vaccination regimen (boosting versus primary) and duration to collect solicited AE (14 versus 7 days) were different, the AE incidence and severity with ZF2001 observed in this study cannot be directly compared with those observed in the ZF2001 efficacy trial [23]. Noteworthily, fever was defined differently (axillary temperature ≥ 37.3°C in our study, but oral temperature ≥ 38.0°C in other studies [6,21,22]). Although AE incidence with SYS6006 was higher than that with ZF2001, most AEs were grade 1 or 2. Higher AE incidence for mRNA vaccines than for other vaccines was not unexpected, as previously reported [6,21,22,24,25].

Heterologous boosting has been more immunogenic than homologous boosting [6,21,22], has been widely used in mass immunization [26], and demonstrated good effectiveness in real-world [10-12]. Our study also showed that SYS6006 heterologous boosting induced robust WT Nab anamnestic response and cross-variant BA.5 Nab, with a 52- to 83-fold increase in Nab titres after vaccination. Furthermore, SYS6006 demonstrated immunogenicity superiority to ZF2001, which was large because SYS6006 encodes the full-length S protein harbouring Omicron key mutations but ZF2001 contains ancestral RBD.

Coincident SARS-CoV-2 infection after vaccination interfered with the immune response to vaccination and should be excluded. The PPS immunogenicity analysis excluded approximately half the participants, mainly due to SARS-CoV-2 infection before blood sampling. Because the immunogenicity subgroup included sufficient participants, the statistical testing was adequately powered.

After the “zero COVID-19” policy was changed, China experienced two nationwide COVID-19 outbreaks predominated by BA.5 (from December 2022 to January 2023) and XBB (May and June 2023) [13]. This study started at the early stage of the BA.5 wave, and went through the late stage of the XBB wave, giving us a unique opportunity to assess the efficacy across the BA.5 and XBB pandemic waves. To the best of our knowledge, this is the first trial to assess VE across two outbreaks. The study showed that SYS6006 boosting conferred protection against RT–PCR-confirmed COVID-19. The relative VE was 51.6% or 42.1% from Day 7 or 14 through Month 6, respectively. The relative VE against BA.5 (66.8% or 84.0%) was higher than that against XBB (37.7% or 30.4%), 7 or 14 days after vaccination, respectively. This was mostly driven by the enhanced immune escape of XBB, and also partially caused by the waning immunity over time. The relative VEs against XBB.1.9 and its sub-lineages, and XBB.1.31 were negative. This was possibly attributed to no sufficient cases accrued for XBB.1.9 and its sub-lineages, and XBB.1.31, but did not indicate any clinical relevance.

Initiated at the beginning of the outbreak, this study provided a unique opportunity to assess the performance of emergency immunization. mRNA vaccine boosting took effect quickly post-Day 7. However, the study timing incurred excessive infected participants before Day 7, therefore reducing case accrual in the PPS population.

Although reliable estimation of the infection rate in the general population during the BA.5 outbreak was unavailable, the infection rate (32.7%) in this study was definitively lower than that in the general population. This is not surprising because the study participants received boosting and most of them had received 3 vaccinations previously which were different from the general population. Per the case definition, only part of the infected participants could be RT–PCR-confirmed cases because of the time window from symptom onset to sampling. The PPS population included fewer cases because of additional requirements.

164 (39%) participants experiencing suspected symptoms during Day 7–180 were not RT–PCR sampled, which resulted in missed potential cases and might decrease surveillance sensitivity. However, the missing proportions between the two groups were similar and, therefore unlikely to biasthe relative efficacy results.

The case surveillance was driven by suspected symptoms. Because SARS-CoV-2 infection-associated symptoms were hard to distinguish from vaccination-associated systemic AEs based on manifestation only, an antigen test was done first for screening cases. If the antigen test was positive, RT–PCR sampling would be done. This was to avoid unnecessary RT–PCR sampling triggered by vaccination-associated systemic AEs. The antigen test was particularly required during the BA.5 wave when vaccination-associated systemic AEs were frequent, but not mandatorily required during the XBB wave. Therefore, not all the RT–PCR positive cases had a preceding positive antigen test, and antigen test was not for the diagnosis of COVID-19 cases. In the post hoc analysis, the relative efficacies against antigen-positive COVID-19 were around 50%. These results were more relevant to the BA.5 wave because most antigen tests were done during this period.

It is ethically challenging to conduct a placebo-controlled trial during a pandemic when vaccines for boosting had been licensed. Therefore, this trial was active-controlled. To demonstrate VE in active-controlled trials sets a higher bar, which theoretically underestimates VE due to the protection conferred by active control. The active control ZF2001 had an efficacy of 75.7% in a placebo-controlled trial [23]. Therefore, a higher efficacy for SYS6006 boosting could be expected in a placebo-controlled trial, which is supported by a report that heterologous boosting with another recombinant protein vaccine V-01 showed an efficacy of 47.8% in a placebo-controlled trial, and an 11.3-fold increase in WT Nab titre on Day 14 [27]. Similarly, ZF2001 had an 11.8-fold increase in WT Nab titre in this study.

Post hoc efficacy analysis included participants who were vaccinated and got the first infection (hybrid immunity) before COVID-19 case accrual, i.e. case confirmation upon a second infection. The relative vaccine efficacy against a second infection-caused COVID-19 was higher than that against the first infection (68.4% versus 46.6%, against XBB), indicating that hybrid immunity of SYS6006 vaccination and infection conferred better protection against reinfection than ZF2001 did. These results were consistent with previous reports that hybrid immunity mounted better immune responses and protection against reinfection [28–30].

Compared with other large-scale phase 3 efficacy trials [23,27,31–34], this trial demonstrated efficacy in a smaller sample size. The coincident COVID-19 peak during the study made it possible to accrue sufficient cases in such a sample size although some cases were missed. Furthermore, there was sufficient difference in COVID-19 incidences between the two groups, hence sufficient power to demonstrate efficacy.

The study was open-label. The control vaccine could be administered only in routine vaccination clinics, but not be taken out for blindly labelling per the official policy. SYS6006 could not be administered in the routine vaccination clinics. The two administration locations were approximately 30 -metres away only. Both were affiliated to The First Hospital of Hebei Medical University. To minimize potential bias, some measures were taken. First, the participants received random numbers blindly. After randomization and immediately before vaccination, the treatment assignment could be disclosed. Second, the investigators responsible for disclosing treatment assignments and injections did not attend other works. Third, antibody assay, RT–PCR test and COVID-19 cases follow-up were done blindly.

Conclusion

SYS6006 heterologous boosting demonstrated good safety, superior immunogenicity and high efficacy against BA.5-associated COVID-19 in the participants who had received two or three doses of SARS-CoV-2 inactivated vaccine, and protected XBB-associated COVID-19, particularly in the hybrid immune population.

Acknowledgements

We thank all the participants in the trial and our external service providers (contract research organizations, laboratories, clinical suppliers, and biostatistics providers) for their assistance.

Disclosure statement

Li CL, Yang HY, Qiu YZ, Zhong X, Ji QL, Zhou FF, Liu KL, Ji L, Wu SQ are the full-time employees of CSPC Megalith Biopharmaceutical Co., Ltd. Li CL and Yang HY are stock owners. All other authors declare no competing interest.

Additional information

Funding

This work was supported by CSPC Megalith Biopharmaceutical Co. Ltd.