Summary
Background The inactivated whole-virus vaccine CoronaVac (SinoVac) is the COVID-19 vaccine most administered worldwide. However, data on its immunogenicity and reactogenicity to heterologous boosting with mRNA vaccines are lacking.
Methods In a cohort of hospital staff in Jakarta, Indonesia, who received two-dose CoronaVac six months prior (median 190 days, IQR165-232), we measured anti-Spike IgG titers on paired serum samples taken before and 28 days after a 100μg mRNA-1273 (Moderna) booster. We performed correlations and multivariable ordinal regressions.
Findings Among 304 participants, the median age was 31 years (range 21-59), 235 (77.3%) were women, 197 (64.8%) had one or more previous SARS-CoV-2 infections (including 155 [51.0%] who had a post-CoronaVac breakthrough infection. Pre-boost IgG titers correlated negatively with the time since the latest documented “virus exposure” (either by the second CoronaVac or SARS-CoV-2-infection whichever most recent). Previous SARS-CoV-2 infection and a longer time interval between second vaccine and mRNA-1273 boost were associated with a higher pre-boost IgG titer. Post-booster, the median IgG titer increased 9.3-fold, from 250 (IQR32-1389) to 2313 (IQR1226-4324) binding antibody units (BAU/mL) (p<0.001). All participants, including seven whose pre-boost IgG was below assay detection limits, became seropositive and all reached a substantial post-boost titer (≥364 BAU/mL). Post-boost IgG was not associated with pre-boost titer or previous SARS-CoV-2 infection. Booster reactogenicity was acceptable, with 7.9% of participants experiencing short-lived impairment of activities of daily living (ADL).
Interpretation A heterologous, high-dose mRNA-1273 booster after two-dose CoronaVac was highly immunogenic and safe, including in those most in need of improved immunity.
Funding Wellcome Trust, UK
Evidence before this study The inactivated whole-virus vaccine CoronaVac (SinoVac) is the COVID-19 vaccine most administered worldwide, at around 2 billion doses in 54 countries. Concerns that CoronaVac has lower immunogenicity than virus vector or mRNA vaccines, with pronounced decreases of neutralising antibody titres within a few months, and reduced effectiveness in the older population, highlight the urgent need for immunogenic, safe and well-tolerated booster schedules, especially with Omicron rapidly emerging.
We used the terms “SARS-CoV-2”, “COVID-19”, “vaccine”, “booster” to search PubMed and medRxiv up to Dec 22th, 2021, with no language or date restrictions, to identify clinical trials and real-world studies reporting on the immune responses and reactogenicity to a “third booster” of currently approved COVID-19 vaccines. Previous research reported that neutralising antibody responses elicited by all currently approved vaccines (mRNA, adenovirus-vectored, inactivated, and protein subunit) declined to varying degrees after 6-8 months after full-schedule vaccination. Several clinical trials have evaluated heterologous (“mix and match”) vaccination schedules, demonstrating robust immune responses in adults. After two-dose CoronaVac, BNT162b2 (Pfizer-BioNTech) boost was significantly more immunogenic than a homologous booster against wild-type and Variants of Concern (VOCs) Beta, Gamma and Delta, and AZD1222 boost increased spike RBD-specific IgG 9-10-fold, with high neutralizing activity against the wild type and VOCs. Compared to previous SARS-CoV-2 variants, current vaccine boosters appeared to neutralise Delta to a slightly lesser degree, and Omicron to a substantially lesser degree, although preliminary data from Moderna found that the authorised dose (50μg) of the mRNA-1273 boost increased antibodies 37-fold and the high-dose (100μg) boost 83-fold.
Added value of this study To our knowledge, this study is the first to provide critical real-world evidence that heterologous boosting with high-dose mRNA-1273 vaccine after CoronaVac is highly immunogenic, safe and well-tolerated in adults. After a primary course of two-dose CoronaVac, we found that a high-dose (100μg) mRNA-1273 booster was immunogenic for all participants in a highly exposed cohort of hospital staff in Jakarta, Indonesia, in the context of Delta predominance, particularly for those with the lowest pre-boost antibody levels. All participants became seropositive and all reached a substantial post-boost titer (≥364 BAU/mL), up to a median 9.3-fold increase. Booster reactogenicity was acceptable, with 7.9% of participants experiencing short-lived impairment of activities of daily living
Implications of all the available evidence The study findings contribute to informing policy makers on flexible options in deploying COVID-19 vaccines in mix-and-match schedules, with particular relevance for countries that are largely dependent on inactivated vaccines. Further trials are warranted that assess clinical endpoints of optimized doses of mRNA-1273 booster, and variant-specific or multivalent vaccines in response to decreased protection against emerging SARS-CoV-2 VOCs.
Introduction
Vaccine-induced population immunity is a key global strategy to control the COVID-19 pandemic, and to date eight COVID-19 vaccines have received Emergency Use Listing (EUL) by the WHO1. Although most studies suggest well preserved protection against severe COVID-19 disease and death, including against variants such as the Delta (B.1.617.2) strain, accumulating evidence shows a progressive decline in protection after a two-dose vaccine schedule in preventing breakthrough infections associated with diminishing humoral immunity over time2–5. In addition, neutralisation and vaccine effectiveness after two-dose vaccine schedules are greatly reduced for the recently emerged Omicron variant (B.1.1.529)6, with significant restoration after a third “booster dose” of an mRNA vaccine7. Recent trials suggested that heterologous (or “mix and match”) virus-vectored or mRNA booster strategies were more immunogenic than a homologous schedule8–12, albeit with increased reactogenicity in some combinations13.
CoronaVac (SinoVac Life Sciences Co Ltd, Beijing, China), an inactivated whole-virus vaccine, received WHO EUL on June 1st, 202114, and is currently the COVID-19 vaccine with most doses administered worldwide at around 2 billion in 54 countries, as per December 20st, 2021. Vaccine efficacy against symptomatic illness of two-dose CoronaVac was reported between 50.5% and 83.5% in trials14,15, and 65.9% and 49.4% in real-world studies in Chile16and Manaus, Brazil, respectively17. Real-world data from Chile reported high prevention of hospitalisation (87.5%), ICU admission (90.3%), and COVID-19-related death (86.3%)16. However, recent studies reported CoronaVac to have lower immunogenicity than virus vector or mRNA vaccines18,19, and a pronounced decrease of neutralising antibody titres within a few months19–21, and reduced effectiveness in the elderly 22,23. This lack of protection is concerning, especially given the rapid emergence of Omicron6.
Indonesia, with the world’s fourth largest population (270 million) has the second highest number of COVID-19 cases (6.26 million) and deaths (144.000) in Asia (per December 20th, 2021)24. The countrywide mass vaccination campaign, launched in January 2021, has been predominantly based on two-dose CoronaVac. The increased transmission of Delta variant during Indonesia’s second COVID-19 surge in July 202125, including among dual-vaccinated health care workers, prompted the Ministry of Health in August 2021 to offer a heterologous, high-dose (100μg) booster of the mRNA-1273 vaccine (Moderna Inc., Cambridge, US) to frontline health care workers.
Given the dearth of data on effectiveness, immunity and reactogenicity of a heterologous booster dose for inactivated whole-virus COVID-19 vaccines18,26, we investigated the immunogenicity and reactogenicity of a high-dose (100μg) heterologous mRNA-1273 booster given six months after two-dose CoronaVac to heavily SARS-CoV-2 exposed hospital staff, measuring anti-spike (anti-S) IgG antibody titers before and one month after the booster, and the occurrence and severity of adverse events within seven days.
Methods
Design and population
The Indonesia Vaccine Immunity & Infection Evaluation (INVITE) study is a longitudinal observational cohort of hospital staff aged 18 years or older in St Carolus and Pasar Minggu hospitals in Jakarta who received the 100μg mRNA-1273 booster dose. The present analysis includes all study participants for whom a paired pre-and post-booster serum sample was available. The study was approved by the research ethics committee of Faculty of Medicine Universitas Indonesia (841/UN2.F1.ETIK/PPM.00.02/2021) and Oxford Tropical Research Ethics Committee (22-21). All participants provided written informed consent.
Procedures
Basic demographic and clinical data were captured on an online case report form. Solicited signs of local or systemic reactogenicity during the seven days following the booster dose were recorded in a daily patient diary, and, after study physician verification, were graded for severity (mild, moderate, severe, or disrupting activities of daily living [ADL]). Past COVID-19 was defined as a previously documented PCR-confirmed SARS-CoV-2 infection. Venous blood samples were drawn on the day of the booster dose (between August 5th, and October 15th, 2021) and 28 days (allowed window +10 days) thereafter (between September 13th and November 12th, 2021).
Serum was stored at -80°C and transferred to a commercial laboratory, where titers of IgG antibodies against the SARS-CoV-2 spike receptor-binding domain were determined using the chemiluminescent microparticle immunoassay (CMIA) SARS-CoV-2 IgG II Quant assay (Abbott Laboratories, Abbott Park, IL, US) on the Architect i2000sr platform, in accordance with the manufacturer’s package insert. Samples with results above the upper limit of quantification were tested again after dilution. The strength of the response (in relative light units) was determined relative to an IgG concurrent calibrator, and reflects the quantity of IgG antibodies present. We expressed test results in WHO International Standard binding antibody units (BAU)/mL using the manufacturer’s conversion factor (1 BAU/mL = 0.142 x arbitray units[AU]/mL)27. Seropositivity was defined as ≥7.1 BAU/mL.
Statistical analysis
Paired comparisons of IgG titers before and after mRNA-1273 booster dose were performed using the Wilcoxon matched pairs signed-rank test. Correlations of log10 IgG titers were expressed using Spearman coefficient (rs). Multivariable ordinal (proportional odds) logistic regression was performed to assess factors associated with pre-and post-boost log10-transformed IgG titer (caterogised in quartiles). Multivariable logistic regression was performed to assess factors associated with occurrence of any severe or disrupting adverse reactions. The independent variables included in the analysis were age, sex, previous SARS-CoV-2 infection, timing of previous SARS-CoV-2 infection (before or after two-dose CoronaVac), presence of any comorbidity (i.e. cardiovascular disease including hypertension, obesity [BMI>30 kg/m2], diabetes mellitus, asthma), pre-boost IgG titer, time interval between first and second CoronaVac dose, time interval between second CoronaVac and mRNA-1273 booster dose. A two-sided P <0.05 was considered significant. Statistical analysis was performed with Stata/IC 15.1 (StataCorp, College Station, TX, USA).
Role of the funding source
The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to the study data.
Results
Participants’ characteristics
There were 304 (86.1%) of 353 cohort participants with complete pre-and post-booster sample pairs, who were included in the analysis. The median age was 31 years (IQR, 27-44, range 21-59), 235 (77.3%) were women, and 28 (9.2%) had one or more comorbidities, including obesity, cardiovascular disease, diabetes mellitus, and asthma (Table 1). There were 197 (64.8%) with previous SARS-CoV-2-infection, including 21 (6.9%) with two infections, and 155 (51.0%) breakthrough infections post-CoronaVac. The mRNA-1273 booster was given at a median of 190 days (IQR 165-232) after the second CoronaVac dose. After the mRNA-1273 booster, the overall median IgG titer showed a 9.3 fold increase, from 250 (IQR 32-1389) to 2313 (IQR 1226-4324) BAU/mL (p< 0.001), and seropositivity increased from 96.4% (293) to 100% (304). (Table 1). Table S1 and Figures S2-3 summarize the pre-post change in median IgG titers for participant subgroups.
Correlations of pre and post mRNA-1273 booster IgG titers
Pre-boost IgG titers varied from 0.97 BAU/mL to 21 466 BAU/mL (4.35 log10 range), and were higher overall in previously infected than in naïve participants (Figure 1 and Figure S1). Pre-boost IgG titers correlated negatively with the time since the most recent documented “virus exposure”, either by the second CoronaVac or SARS-CoV-2 infection (whichever most recent). Participants with a breakthrough infection after post-Coronavac had the shortest time since most recent “virus exposure”, and had a higher pre-boost IgG titer than those infected before CoronaVac and naïve participants. Participants with a breakthrough infection post-CoronaVac had the shortest time since most recent “virus exposure” and a higher pre-boost IgG titer than the naïve participants.
All participants reached a substantial post-boost titer, ranging between 364 and 195 452 BAU/mL (2.73 log10 range), and this was irrespective of the pre-boost IgG titer, previous infection, or the timing of the previous infection (either before or after two-dose CoronaVac vaccination). Participants with a breakthrough infection after two-dose CoronaVac had a higher pre-boost IgG titer than those infected before Coronavac and the naïve participants. Post-boost IgG titers were in the same range for all three subgroups (Figure 3). Post-boost IgG titers did not differ for previous SARS-CoV-2 infection, age, sex, and presence of any comorbidity (Figure S1 and S2).
Factors associated with pre- and post-booster IgG titers
In multivariable analysis, previous SARS-CoV-2 infection either before or after two-dose CoronaVac, and a longer time interval between second and mRNA-1273 boost dose were associated with a higher pre-boost IgG titer (p<0.0001 each); age, sex, comorbidity, time interval between first and second dose were not associated. None of these factors, including pre-boost IgG titer, were associated with a higher post-booster IgG titer (Table 2).
Reactogenicity of mRNA-1273 booster
The 304 participants reported a total of 300 adverse reactions within 7 days after mRNA-1273 booster. The percentage of participants with any solicited adverse reaction was 98.7% (300), 96.4% (293) for any local reaction (injection site pain and swelling most frequent) and 90.8% (276) for any systemic reaction (myalgia, fever/chills and headache being most frequent). Most solicited local and systemic adverse reactions were mild (86, 28.3%) or moderate (131, 43.1%), whereas severe (59, 19.4%) or those disrupting ADL (24, 7.9%) were less common (Figure 4 and Table S2). All adverse reactions were short-lived and none required hospitalisation. Severe or ADL-disrupting adverse reactions were associated with longer interval between second dose and mRNA1273 booster (OR 1.37 per month increase, 1.09-1.72, p=0.001), and inversely associated with age (OR 0.71 per decade increase, 0.54-0.92, p=0.011); sex, comorbidity, previous SARS-CoV-2 infection, time interval between first and second dose, and pre- and post-boost IgG titer were not associated with any adverse reactions (Table S3).
Discussion
This real-world study found that a heterologous mRNA-1273 vaccine boost dose in Indonesian hospital staff with high SARS-CoV-2 exposure, after a primary course of two-dose CoronaVac six months prior, resulted in a median 9.3-fold increase in anti-spike serum IgG titers measured after 28 days, from a median of 250 (IQR 32-1389) to 2313 (IQR 1226-4324) BAU/mL. All participants, including seven individuals whose pre-boost IgG was below assay detection limits, became seropositive and all reached a substantial post-boost titer (≥364 BAU/mL). Post-boost IgG levels were not associated with pre-boost antibody titer or previous SARS-CoV-2 infection. Higher pre-boost IgG levels were associated with documented previous SARS-CoV-2 infection, and, unexpectedly, a longer interval since the second vaccine dose. However, when we assessed time since the most recent documented “virus exposure” (either by vaccine or infection), the pre-boost IgG titers correlated negatively, as expected19–21. These findings, coupled with the high rate of documented post-CoronaVac virus breakthroughs, confirm intense SARS-CoV-2 exposure among front line heath care workers during the Delta surge in July 202125. Delta breakthrough infections have been associated with more potent, durable and resilient responses to spike mutations observed in variants of concern than those observed in subjects who were infected only or received only two doses of COVID-19 vaccine28.
The reactogenicity of the high-dose heterologous mRNA-1273 booster was acceptable, with common reports of injection site pain and swelling as well as muscle pain, fever/chills and headache, and 7.9% experienced a short-lived impairment in performing ADL. The booster was better tolerated with increasing age and a shorter time since the second dose. The findings largely concur with Moderna’s phase 2/3 trial of the high-dose (100μg) booster dose, which was generally safe and well-tolerated, although with slightly more frequent adverse reactions compared to the authorised 50μg dose29.
Emerging evidence for currently available vaccines suggests that heterologous boosting improves the breadth of humoral and cellular protection more than homologous boosting11,12. A small randomized trial in Hong Kong in adults who had received two-dose CoronaVac with low antibody response, found that an additional heterologous mRNA vaccine booster dose of BNT162b2 (Pfizer-BioNTech) was significantly more immunogenic than a homologous booster dose of CoronaVac booster against wild-type and VOCs18. Another study in heath care workers in Thailand who received two-dose CoronaVac found that a heterologous virus-vector booster AZD1222 (AstaZeneca) increased spike RBD-specific IgG 9-10-fold, with high neutralizing activity against the wild type and VOCs26.
Policy makers in several countries have started implementing third or periodic boosting to protect the most vulnerable populations, and mitigate health care and economic impacts. Real-world data and cohort studies are critical to guide decisions regarding when, which populations and what boosters should be administered. Compared to previous SARS-CoV-2 variants, current vaccine boosters appeared to neutralise Delta to a slightly lesser degree12, and Omicron to a substantially lesser degree7. Nonetheless, on 12th December, Moderna announced preliminary data that the authorised booster dose of 50μg mRNA-1273 increased neutralising antibodies 37-fold and the high-dose of 100μg 83-fold, compared to pre-boost levels29. Further trials on clinical endpoints of optimized booster doses and variant-specific or multivalent vaccines in response to decreased susceptibility to neutralization of emerging SARS-CoV-2 VOCs are urgently needed30.
There are some limitations to our study. First, although accumulating evidence suggests that anti-Spike IgG response is a correlate of disease protection31,32, it is important to recognise there is not yet an established or well defined correlate of long-term vaccine protection. In this study we did not assess anti-N IgG, neutralizing antibody or cellular immunity33. Second, because subjects had not been under routine surveillance for SARS-CoV-2 infection, we could not fully discern all who had been infected by SARS-CoV-2, and the use of anti-N as a proxy for breakthrough infections was not valid because of the initial immunization with a whole-virus inactivated vaccine. Third, the sample size and follow-up period were not sufficient to identify less common or late adverse events following booster vaccination, and the immunogenicity data were limited to immune responses through study day 28. Fourth, only immunological data were collected and therefore this investigation lacks information regarding the efficacy of a heterologous mRNA-1273 booster vaccination. Lastly, demographics of the volunteers were not representative of the Indonesian population, and elderly were not represented.
In conclusion, our study provides evidence that, after a primary course of CoronaVac inactivated virus vaccine, a heterologous, high-dose mRNA-1273 booster was highly immunogenic and safe, even for those with the lowest pre-boost antibody levels.
Data Availability
After publication, the datasets used for this study will be made available to others on reasonable requests to the corresponding author, including a detailed research proposal, study objectives and statistical analysis plan. Deidentified participant data will be provided after written approval from the principal investigators.
Declaration of interests
SJD is a scientific advisor to the Scottish Parliament on COVID-19 immunology, for which she receives a fee. Oxford University through EOCRU received funding from SinoVac LifeSciences Ltd, Beijing, China for a vaccine booster trial, unrelated to the current study. All other authors declare that they have no conflict of interest.
Funding
This work was funded by the Wellcome Trust, UK (106680/Z/14/Z and 222574/Z/21/Z). SJD is funded by an NIHR Global Research Professorship (NIHR300791).
Authors’ contributions
EN and RLH are the INVITE principal investigators. RS, DU, EN, AS and RLH conceptialised the study. RS, DU, CN, F, DT, MR, AFR, established the cohort, collected the clinical samples and data. S and DS designed, supervised and performed the antibody testing. NN and WH managed the clinical database. HS performed the statistical analysis, with input from AS and RLH. RLH wrote the first draft of the manuscript. All authors provided valuable input to interpretation of the data and critically reviewed the paper and figures for important intellectual content. All authors reviewed and approved the final version of the manuscript.
Data sharing statement
After publication, the datasets used for this study will be made available to others on reasonable requests to the corresponding author, including a detailed research proposal, study objectives and statistical analysis plan. Deidentified participant data will be provided after written approval from the principal investigators.
Acknowledgements
The authors thank the study participants, the staff at the collaborating clinical sites and the Prodia laboratory.