Enhanced immunity against SARS-CoV-2 in returning Chinese individuals

ABSTRACT

Global COVID-19 vaccination programs effectively contained the fast spread of SARS-CoV-2. Characterizing the immunity status of returned populations will favor understanding the achievement of herd immunity and long-term management of COVID-19 in China. Individuals were recruited from 7 quarantine stations in Guangzhou, China. Blood and throat swab specimens were collected from participants, and their immunity status was determined through competitive ELISA, microneutralization assay and enzyme-linked FluoroSpot assay. A total of 272 subjects were involved in the questionnaire survey, of whom 235 (86.4%) were returning Chinese individuals and 37 (13.6%) were foreigners. Blood and throat swab specimens were collected from 108 returning Chinese individuals. Neutralizing antibodies against SARS-CoV-2 were detected in ~90% of returning Chinese individuals, either in the primary or the homologous and heterologous booster vaccination group. The serum NAb titers were significantly decreased against SARS-CoV-2 Omicron BA.5, BF.7, BQ.1 and XBB.1 compared with the prototype virus. However, memory T-cell responses, including specific IFN-γ and IL-2 responses, were not different in either group. Smoking, alcohol consumption, SARS-CoV-2 infection, COVID-19 vaccination, and the time interval between last vaccination and sampling were independent influencing factors for NAb titers against prototype SARS-CoV-2 and variants of concern. The vaccine dose was the unique common influencing factor for Omicron subvariants. Enhanced immunity against SARS-CoV-2 was established in returning Chinese individuals who were exposed to reinfection and vaccination. Domestic residents will benefit from booster homologous or heterologous COVID-19 vaccination after reopening of China, which is also useful against breakthrough infection.

KEYWORDS:

• COVID-19
• SARS-CoV-2
• returning Chinese individuals
• Omicron
• neutralizing antibody
• memory T-cell response

Introduction

The coronavirus disease 2019 (COVID-19) pandemic caused by the novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has continued for three years since the beginning of 2020. As of April 2023, over 0.7 billion cases and 7 million deaths have been recorded related to COVID-19 according to World Health Organization (WHO) reports.¹ The socioeconomic threat and burden of COVID-19 worldwide are immeasurable.² A vaccine against SARS-CoV-2 was considered the key to containing the global spread of COVID-19. A number of COVID-19 vaccines (CoronaVac, BNT162b2, mRNA-1273, Sputnik V and AZD1222) have been developed and launched at an incredible accelerated speed versus traditional vaccine development programs. Indeed, COVID-19 vaccination was shown to be effective in preventing infection in individuals and reducing severe disease and death^3–5 until the emergence of SARS-CoV-2 variants.

SARS-CoV-2 has evolved dynamically and resulted in a number of novel variants of concern (VOCs), including B.1.1.7 (Alpha) in southeast England,⁶ B.1.351 (Beta) in South Africa,⁷ B.1.525 (Eta) in Nigeria,⁸ P.1 (Gamma) in Brazil,⁹ Delta (B.1.617.2) in India, and Omicron.^10–12 For the most recent Omicron lineage, breakthrough infections are encountered more frequently than with any other VOCs due to substantial evolution and consequent immune evasion to available vaccines.^13,14 For instance, three-dose vaccination is effective against Delta infection but less effective against Omicron infection.¹⁵ Although two doses of the BNT162b2 vaccine are associated with high protection against SARS-CoV-2 infection within 6 months,¹⁶ primary immunization alone is not enough to protect individuals from breakthrough infection by multiple VOCs. Booster doses are currently used against COVID-19 either in homologous or heterologous vaccination and have been evaluated as safer and more protective in preventing VOCs.^17,18 During the BA.4- and BA.5-predominant period, vaccine preventive efficiency associated with protection against medically attended COVID-19 illness was lower with increasing time since the last dose; estimated vaccine preventive efficiency was higher after the receipt of 1 or 2 booster doses compared with a primary series alone.¹⁹ Regarding Omicron, reboosting with a third or fourth dose is effective in protecting individuals against severe COVID‐19‐related outcomes.^20–22 These are the effects of vaccination against Omicron breakthrough infections. However, an observational study in Italy showed that simultaneous administration of a quadrivalent influenza vaccine and an mRNA anti-SARS-CoV-2 vaccine neither affected the safety of the seasonal influenza vaccine nor was associated with a higher risk of SARS-CoV-2 breakthrough infection.²³

The public health strategies implemented against COVID-19 were variable depending on the socioeconomic level and policies of public health authorities of each country. In China, the community-based dynamic zero-COVID policy was the key to preventing endemic COVID-19 from the beginning of 2020 to the end of 2022, in which a strict “14 + 7” in the early stage and an adapted “7 + 7” quarantine policy (14 or 7 days in a quarantine station plus 7 days of quarantine at home) intercepted most COVID-19 cases in China. Thus, the socioeconomic burdens of COVID-19 were minimized in China. In contrast, reopening policies after first-wave lockdowns were evaluated in the USA and European countries as early as the end of 2021. An study in 22 European countries reported that reopening policies were associated with a 1.5-percentage point increase in mobility and a 4% increase in subsequent infections after 2 weeks.²⁴ To reduce imported cases, Australia implemented mandatory 14-day quarantine, with testing within 48 h after arrival and between days 10 and 12 of quarantine.²⁵ Although these multiplex tests overseas could quickly screen for cases, the strategy did not include any measures to reduce the burden of a long quarantine or contain cases of asymptomatic transmission over the past years.²⁶ In China, a comment from a scientist in 2022 suggested termination of the dynamic zero-COVID policy and reopening based on rapid vaccination rollout and demand for socioeconomic development.²⁷ Certainly, this delayed reopening led to completely different basic immunity statuses of residents between COVID-19-endemic countries and China, and the herd immunity obtained through sole vaccination may be significantly weaker than that of foreigners who experienced complicated vaccination procedures and breakthrough infections. Indeed, a host serological survey in Wuhan reported 3.2% ~ 3.8% positive rates from March to April 2020,²⁸ whereas a survey among residents in Chelsea, USA, found that 32% of those were SARS-CoV-2 antibody positive.²⁹ However, people with asymptomatic infections had low positivity in early IgG-IgM serologic testing but a high viral load in RT‒PCR testing, suggesting that serologic testing does not accurately indicate SARS-CoV-2 infection in asymptomatic infections (IgM and IgG sensitivity of 27.78% and 50.00%, respectively).^30,31 Therefore, there is a need to combine serological testing with immunization to evaluate the immunity status of returned populations from COVID-19-endemic countries and understand the achievement of herd immunity and long-term protection after reopening in China. In this study, a survey was conducted among returning travelers in 7 quarantine stations in Guangzhou, China, before reopening was implemented in December 2022. Demographic characteristics, vaccination status, infection history, serum immunity characteristics, and associated influencing factors were evaluated accordingly.

Materials and methods

Study design

This study was conducted at 7 quarantine stations in Guangzhou, China, from December 2022 to March 2023. All returning travelers who provided informed consent were included in this study according to the following criteria: 1) age ≥18 years; 2) negative SARS-CoV-2 RNA detected by real-time RT‒PCR; 3) normal physical health; and 4) travel history of more than 1 month. The exclusion criteria were immunocompromised patients and recipients of blood and blood products during the last 6 months. This study was approved by the institutional review committee of the Guangdong Provincial Center for Disease Control and Prevention.

Demographic and epidemiological information collection

The information of each subject was collected by questionnaire survey, and the investigation content was as follows: 1) basic information: sex, age, ID number, and telephone number; 2) health status and lifestyle: height, weight, chronic medical history, smoking, alcohol consumption and physical activity; 3) SARS-CoV-2 vaccination, including vaccine brands, doses and time of vaccination; and 4) COVID-19 history.

Specimen collection and storage

Blood and throat swab specimens were collected from each participant. Venous blood and throat swabs were transported to the laboratory of the Guangdong Provincial Institute of Public Health at 4°C, and whole blood was transported at room temperature (RT). On the day of collection, peripheral blood mononuclear cells (PBMCs) were separated from whole blood by using the Lymphocyte Separation Tube for Human Peripheral Blood (DAKEWE, Shenzhen, China). Venous blood was centrifuged at 3000 rpm/min for 5 min (Thermo Fisher Scientific, Dreieich, Germany), and the serum was transferred into new 2-mL cryogenic vials. The serum and throat swab samples were frozen at −80°C until testing, and the serum samples were inactivated at 56°C for 30 min before the assay.

Nucleic acid extraction and rRT-PCR

Throat swab specimens were treated as previously described.³² Total RNA was extracted using a virus nucleic acid preload kit (Magen Biotechnology Co., Ltd., Guangzhou, China). SARS-CoV-2 RNA was analyzed with a commercial rRT-PCR kit (DaAn Gene, Guangzhou, China) targeting the ORF1ab and nucleocapsid (N) genes. Amplification was performed on an Applied Biosystems™ 7500 real-time fluorescent PCR instrument (Thermo Fisher Scientific, Waltham, MA, USA). A CT value < 40 was considered SARS-CoV-2 RNA positive in throat swab samples.

Determination of N and S proteins (S1 RBD) by enzyme-linked immunosorbent assay (ELISA)

The SARS-CoV-2 N and S proteins (S1 RBD) were detected in serum by ELISA (RayBiotech, Peachtree Corners, GA, USA) (Cat. no. ELV-COVID19N, ELV-COVID19S1) according to the manufacturer’s protocol. The serum samples were initially diluted at a ratio of 1:6. One hundred µL of each standard and sample was added into appropriate wells coated with antivirus COVID-19 N protein or S protein (S1 RBD). The wells were covered and incubated for 2.5 hours at RT with gentle shaking. The solution was discarded, the cells were washed 4 times, and then 100 µL of biotinylated antibody was added to each well. The cells were incubated for 1 hour under the same conditions. Then, the solution was discarded, and the wash was repeated as before. One hundred µL of streptavidin solution was added, and the solution was incubated for 45 minutes. After that, the solution was discarded, and the wash was repeated as before. TMB One-Step Substrate Reagent (100 µL) was added to each well and incubated for 30 minutes. Fifty µL of Stop Solution was added to each well. All ELISA results were determined using an enzyme-labeled instrument (BioTek Epoch, Winooski, VT, USA), and the data are presented as the OD (optical density) value at the wavelength λ = 450 nm. The OD value was converted to concentration values (N, ng/mL; S, pg/mL) using SigmaPlot software. The original values of the N protein >0.07 ng/mL and S protein >4.5 pg/mL were considered positive. The final values were multiplied by 6 dilution factors before analysis.

Competitive enzyme-linked immunosorbent assay (ELISA)

The neutralizing antibody (NAb) in serum was detected using a commercial ELISA anti-SARS-CoV-2 S kit (Shanghai GeneoDx Biotech, Ltd., Co., Shanghai, China). The serum samples were initially diluted at a ratio of 1:100. The experimental procedure was conducted strictly according to the kits’ manual instructions as previously described.³³ All ELISA plates were evaluated by using an enzyme-labeled instrument (BioTek Epoch, Winooski, VT, USA), and the data are presented as the OD (optical density) value at the wavelength λ = 450 nm. The OD value was converted to antibody concentration values (IU/mL) using software from Shanghai GeneoDx Biotech. Original values over 6.5 IU/mL were used as a positive threshold. The final values were multiplied by the dilution factor before analysis.

Microneutralization assay with SARS-CoV-2 variants

Prototype SARS-CoV-2 (2020XN4276) and Omicron sublineages (BA.5, GDPCC 2.00303; BF.7, GDPCC 2.01504; BQ.1, GDPCC 2.01502; XBB.1, GDPCC 2.01503) of SARS-CoV-2 were isolated from confirmed COVID-19 cases in the laboratory of the Guangdong Provincial Center for Disease Control and Prevention. An end point dilution microplate neutralization assay was performed to measure neutralization activity. The titer of each serum sample was determined according to the standard protocol as previously described.³⁴ Titers of serum samples ≥ 4 were considered a positive threshold.

SARS-CoV-2 enzyme-linked immuno-FluoroSpot assay

To evaluate the memory T-cell (MTC) response to SARS-CoV-2, specific interferon gamma (IFN-γ)- and interleukin-2 (IL-2)-secreting T cells were detected using human IFN-γ/IL-2 SARS-CoV-2 FluoroSpot PLUS kits according to the manufacturer’s protocol (Mabtech AB, Sweden). The plates precoated with capturing monoclonal anti-IFN-γ and anti-IL-2 were incubated overnight in RPMI-1640 medium containing 10% fetal calf serum (FCS) supplemented with a mixture containing the SARS-CoV-2 peptide pool (S1 scanning and SNMO pools), anti-CD28 (5 µg/ml) and 250,000 PBMCs per well at 37°C and 5% CO₂ for 24 h. The anti-CD3 (5 µg/ml) response served as a positive control for polyclonal T-cell activation. PBMCs in culture without the addition of SARS-CoV-2 peptides were used as the negative control. The plates were washed five times with PBS before incubation with diluted 1:200 anti‐IFNγ (7‐B6‐1‐BAM) and 1:500 anti‐IL‐2 (MT8G10‐biotinylated) antibodies for 2 h at RT, followed by 1 h of incubation with secondary fluorophore-conjugated antibodies, anti‐BAM 490 (1:200), and streptavidin‐550 (1:200). Finally, the fluorophore enhancer was added for 10 min. Within each step, the plates were washed five times with PBS, except after the addition of the fluorophore enhancer. The MTC results were evaluated using an IRIS reader (Mabtech AB, Sweden) at wavelengths λ = 490 nm (IFN-γ) and λ = 550 nm (IL-2). The results were expressed as spot forming units (SFU), the number of spots after subtracting the background spots of the negative control. The SFU of each sample were calculated using the means of duplicate wells and expressed as SFU/250,000 PBMCs. The cutoff value was set at the highest number of spots from the negative controls.

Statistical analysis

All data was analyzed by IBM SPSS Statistics 25 software (SPSS Inc., Chicago, IL, USA). The statistical analyses for the authentic virus neutralization assessments were performed using GraphPad Prism 8.0.2 (GraphPad Software, San Diego, CA, USA) for calculation of the mean value for each data point. Each specimen was tested in duplicate. The NAb titers of the prototype virus and Omicron sublineages of SARS-CoV-2 were log2-transformed prior to analysis and compared by the Mann‒Whitney U test. The correlation between the NAb titers of the prototype virus and the Omicron sublineages of SARS-CoV-2 and SFU were measured using Spearman’s rank correlation coefficient.

To analyze the factors affecting the NAb titers, we first described the median of the log2-transformed prior and conducted univariate analyses of all demographic information to select the independent variables that were significantly correlated with the change in the outcome measure. Then, the variables selected from the univariate analyses were included in the multivariate linear regression model. For variables that were categorical, we set dummy variables for treatment. A p < .05 was considered statistically significant.

Results

Demographic information and epidemiological findings

From December 2022 to March 2023, a total of 272 subjects were recruited from 7 quarantine stations, including 235 (87.4%) returning Chinese individuals and 37 (13.6%) foreigners. Most of them were male (210/272, 77.2%). According to their self-report questionnaires (Table 1), 183 (67.3%) were nonsmokers, 27 (9.9%) were ever smokers, and 62 (22.8%) were current smokers. Regarding alcohol usage, 101 (37.1%) were never drinkers, 104 (38.2%) were ever drinkers, and 67 (24.6%) were current drinkers. A total of 194 (71.3%), 69 (25.4%), 8 (2.9%), and 1 (0.4%) had had COVID-19 0, 1, 2 and 3 times, respectively. A total of 21 (7.8%), 2 (0.7%), 49 (18.0%), 166 (61.0%) and 34 (12.5%) had received 0, 1, 2, 3 and 4 doses of a SARS-CoV-2 vaccine, respectively (Table 1).

Table 1. Demographic and epidemiological information of 272 subjects (235 returning Chinese individuals, 37 foreigners).

	N (%)		N (%)
Sex		Smoking status
Male	210 (77.2%)	Never smoker	183 (67.3%)
Female	62 (22.8%)	Ever smoker	27 (9.9%)
		Current smoker	62 (22.8%)
Age (years)	37.77 ± 10.08^a
≤29	52 (19.1%)	Alcohol consumption
30–39	90 (33.1%)	Never drinker	101 (37.1%)
40–49	60 (22.1%)	Ever drinker	104 (38.2%)
≥50	31 (11.4%)	Current drinker	67 (24.6%)
Height (centimeters)	169.44 ± 7.38^a	COVID-19
		Yes	78 (28.7%)
Weight (kilograms)	71.69 ± 18.10^a	No	194 (71.3%)
Medical history of chronic diseases		Number of COVID-19 infections
Yes	14 (5.1%)	0	194 (71.3%)
No	258 (94.9%)	1	69 (25.4%)
		2	8 (2.9%)
Comorbidities		3	1 (0.4%)
Hypertension	6 (2.2%)
Diabetes	4 (1.5%)	SARS-CoV-2 vaccine dose
Cardiovascular disease	1 (0.4%)	0	21 (7.8%)
Other diseases	5 (1.8%)	1	2 (0.7%)
		2	49 (18.0%)
Time of last vaccination		3	166 (61.0%)
2020	5 (1.8%)	4	34 (12.5%)
2021	121 (44.5%)
January-April 2022	61 (22.5%)	Specimen Collection
May-August 2022	33 (12.1%)	Yes	108 (39.7%)
September-December 2022	31 (11.4%)	No	164 (60.3%)
Unknown	21 (7.7%)

^aAge, height, and weight were normally distributed and are presented as the means ± standard errors (SEM).

Among 272 subjects, 135 (49.7%) refused to donate specimens, 24 (8.8%) reported incomplete information, and 5 (1.8%) foreigners were excluded from further analysis. Finally, 108 (39.7%) returning Chinese individuals agreed to specimen collection, including 89 (82.4%) males and 19 (17.6%) females (Supplementary Table S1). All throat swabs tested negative for SARS-CoV-2.

Antigens (N and S proteins) and neutralizing antibodies (competitive ELISA) in serum

Although none of the participants were positive for SARS-CoV-2 RNA, the N protein and S protein of SARS-CoV-2 were determined in the serum by ELISA. The N protein was detected in 19 (17.6%) specimens with a median concentration of 0.97 (0.76 ~ 1.61) ng/mL, while the S protein was not detected in any specimens (Supplementary Figure S1a, Table 2). To assess the antibody responses of 108 participants with different vaccination histories, NAb were initially measured by competitive ELISA. The results showed that the neutralizing antibody concentration in the heterologous booster vaccination group was higher than that in the primary vaccination group (5599.32 IU/mL vs. 1264.16 IU/mL, Z = −2.424, p = .015) and the homologous booster vaccination group (5599.32 IU/mL vs. 899.35 IU/mL, Z = −3.626, p < .001) (Table 2). However, there was no significant difference between the primary vaccination group and the homologous vaccination group (Z = −1.033, p = .302) (Supplementary Figure S1b).

Table 2. SARS-CoV-2 N and S proteins and neutralizing antibodies by ELISA.

	N	Positive (%)	Concentration	IQR
NAb (IU/mL)
primary	17	17 (100.0)	1264.16	518.45 ~ 4525.83
homologous booster	68	38 (100.0)	899.35	251.98 ~ 3673.31
heterologous booster	23	23 (100.0)	5599.32	1392.12 ~ 17855.65
N protein (ng/ml)	108	19 (17.6)	0.97	0.76 ~ 1.61
S protein (pg/ml)	108	0	/

Concentration: The data are shown as the medians; IQR: interquartile range; primary: only one or two doses of COVID-19 vaccine; homologous booster: two inactivated COVID-19 vaccines with an inactivated booster; heterologous booster: two inactivated COVID-19 vaccines with an mRNA/adenovirus vector/recombinant protein booster.

NAb titers in serum against VOCs

We determined the NAb titers of 108 participants against the prototype virus and four Omicron variants, e.g., BA.5, BF.7, BQ.1 and XBB.1. The results showed that ~ 90% of participants’ serum showed positive neutralizing antibodies against the prototype virus, whereas the NAb titers were significantly decreased against the Omicron variants compared to the prototype virus (Figure 1). In the primary vaccination group, the geometric mean titers (GMTs) dropped from 72 for the prototype virus to 20 for Omicron BA.5 (2.95-fold, P = .010), 20 for BF.7 (2.95-fold, P = .016), 7 for BQ.1 (12.19-fold, P < .001), and 5 for XBB.1 (18.75-fold, P < .001) (Figure 1a). In the homologous booster vaccination group, the GMT dropped from 57 for the prototype virus to 12 for Omicron BA.5 (3.83-fold, P < .001), 10 for BF.7 (4.80-fold, P < .001), 5 for BQ.1 (10.60-fold, P < .001), and 4 for XBB.1 (13.50-fold, P < .001) (Figure 1b). In the heterologous booster vaccination group, the GMT dropped from 248 for prototype virus to 40 for BA.5 (5.40-fold, P < .001), 33 for BF.7 (6.76-fold, P < .001), 8 for BQ.1 (31.00-fold, P < .001), and 8 for XBB.1 (31.00-fold, P < .001) (Figure 1c). Surprisingly, we found 9 participants with a higher NAb titer against BA.5 than that against the prototype virus or any other variants. Among them, each participant with primary vaccination and a homologous booster reported a single SARS-CoV-2 infection, while the remaining 7 participants with a homologous booster denied an infection history. Six participants had higher NAb titers against BF.7, and one reported a history of three infections, but the other 5 participants (primary, 1; homologous booster, 2; heterologous booster, 2) denied having an infection history. In addition, 6 participants showed higher NAb titers against both BA.5 and BF.7 than against the prototype virus or any other variants. Each participant with primary vaccination and homologous booster reported a single infection, 2 participants with heterologous booster reported two infections, and the remaining 2 participants (primary, 1; heterologous booster, 1) denied an infection history.

Figure 1. NAb titers of participants with (a) primary, (b) homologous booster and (c) heterologous booster vaccination against the prototype virus and a variety of omicron variants of concern (VOCs).

Neutralizing antibody titers against the prototype virus and 4 Omicron VOCs, BA.5, BF.7, BQ.1, and XBB.1, were tested for in the serum of 108 participants. Positive rate. the number of participants with NAb titers ≥ 4 in the primary/homologous booster/heterologous booster vaccination group; GMT. geometric mean titer.

Memory T-cell responses

We assessed MTC responses in 33 specimens by measuring the number of specific IFN-γ- and IL-2-secreting T cells using the FluoroSpot assay. Using the highest value observed from all of the negative controls as a cutoff, specific IFN-γ-secreting T cells were detected in 75.0% (3/4), 61.5% (16/26) and 33.3% (1/3) of the primary vaccination, homologous booster vaccination and heterologous booster vaccination groups, respectively. Specific IL-2-secreting T cells were detected in 75.0% (3/4), 53.8% (14/26) and 66.7% (2/3) of the primary vaccination, homologous booster vaccination and heterologous booster vaccination groups, respectively. Among these participants, 3 of 4 in the primary vaccination group reported a recent infection history (3/4, 75.0%), 8 of 26 in the homologous booster group reported a recent infection (8/26, 30.8%), and 2 of 3 in the heterologous booster group reported a recent infection (2/3, 66.7%).

The number of specific IFN-γ T cells showed no differences in the primary vaccination group (26 SFU), homologous booster vaccination group (27 SFU) and heterologous booster vaccination group (2 SFU) (primary vs. homologous, Z = −0.428, p = .669; primary vs. heterologous, Z < 0.001, p > .500; homologous vs. heterologous, Z = −0.646, p = .519) (Figure 2a), as well as for the specific IL-2 T cells in statistical assessment (primary, 12 SFU; homologous, 8 SFU; heterologous, 7 SFU) (primary vs. homologous, Z = 0.459, p = .646; primary vs. heterologous, Z < 0.001, p > .500; homologous vs. heterologous, Z = −0.108, p = .914) (Figure 2b). Figure 2c indicates the spots of representative samples in positive, negative, stimulated and blank wells.

Figure 2. SARS-CoV-2-specific memory T-cell (MTC) responses in participants (N = 33) with primary, homologous booster and heterologous booster vaccination. (a) Number of specific IFN-γ-secreting T cells. (b) Number of specific IL-2-secreting T cells. (c) The spots from sample 93.

The highest spot value observed from all of the negative controls was used as a cutoff (IFN-γ, 8; IL-2, 4).

Consistency of NAb titers with NAb/MTC/N protein

The Spearman correlation assay between microneutralization assay NAb titers and competitive ELISA NAb, specifically IFN-γ- and IL-2-secreting T cells and N protein levels showed that NAb titers were more correlated with competitive ELISA NAb (prototype, r = 0.823, p < .001; Omicron BA.5, r = 0.728, p < .001; Omicron BF.7, r = 0.678, p < .001; Omicron BQ.1, r = 0.656, p < .001; Omicron XBB.1, r = 0.595, p < .001) (Supplementary Figure S2a). The number of specific IFN-γ- or IL-2-secreting T cells or N protein levels showed no correlation with NAb titers (Supplementary Figure S2b).

Factors affecting NAb titers

The logarithm of NAb titers was used as the dependent variable, and univariate regression analysis was applied to each variable. The results showed that NAb titers were significantly correlated (P ≤ .05) with smoking, alcohol consumption, COVID-19, SARS-CoV-2 vaccination, and the interval between last vaccination time and the sampling time. The vaccine dose was a common influencing factor for the titers of the five variants (P ≤ .05) (Table 3). Based on the significant correlation factors in univariate analysis, we developed a multiple linear regression model to further analyze their correlations. We found that smoking, vaccination status, alcohol consumption and COVID-19 were significant factors in the multivariate analysis of NAb titers of the prototype virus and four VOCs (P < .05) (Supplementary Tables S2–S5), while the multivariate analysis of NAb titers for Omicron XBB.1 was not significant (Supplementary Table S6).

Table 3. Univariate analysis of factors related to NAb titers for SARS-CoV-2.

	Prototype	Omicron BQ.1	Omicron XBB.1	Omicron BA.5	Omicron BF.7
Sex
Male	0.313 (0.577)	0.104 (0.748)	0.000 (0.991)	0.000 (0.995)	0.106 (0.745)
Female
Age (years)
≤29	1.880 (0.137)	0.297 (0.827)	0.412 (0.745)	0.062 (0.980)	0.334 (0.801)
30–39
40–49
≥50
Height (centimeters)	3.884 (0.051)	0.724 (0.397)	0.194 (0.660)	2.946 (0.089)	2.227 (0.139)
Weight (kilograms)	2.094 (0.151)	0.410 (0.524)	0.011 (0.916)	1.717 (0.193)	0.280 (0.598)
Medical history of chronic diseases
No	0.338 (0.562)	0.005 (0.942)	0.012 (0.913)	0.052 (0.820)	0.068 (0.795)
Yes
Smoking status
Never smoker	3.104 (0.049)	3.419 (0.036)	2.043 (0.135)	4.504 (0.013)	3.845 (0.024)
Ever smoker
Current smoker
Alcohol consumption
Never drinker	1.384 (0.255)	4.287 (0.016)	1.814 (0.168)	3.865 (0.024)	3.535 (0.033)
Ever drinker
Current drinker
COVID-19
No	1.894 (0.172)	4.049 (0.047)	3.139 (0.079)	4.826 (0.030)	3.217 (0.076)
Yes
Number of COVID-19 infections
0	0.762 (0.518)	1.962 (0.124)	2.926 (0.037)	2.466 (0.066)	2.405 (0.072)
1
2
3
SARS-CoV-2 vaccine dose
1	2.912 (0.038)	3.465 (0.019)	4.413 (0.006)	2.775 (0.045)	4.152 (0.008)
2
3
4
Inoculation status
Primary	6.542 (0.002)	1.863 (0.160)	3.336 (0.039)	4.503 (0.013)	5.529 (0.005)
Homologous booster
Heterologous booster
Time interval	4.812 (0.030)	1.741 (0.190)	7.640 (0.007)	4.552 (0.035)	3.772 (0.055)

The values in the table are F values after univariate analysis of the respective variables, and the values in parentheses are P values; p ≤ .05 (two-sided) is considered statistically significant.

The multivariate analysis showed that compared with nonsmokers (median log2 = 7), smokers (median log2 = 6) had a significantly lower NAb titer for the prototype virus (β = −0.216, 95%CI = −2.426~−0.183). Compared with the primary vaccination group (median log2 = 7), the heterologous booster group (median log2 = 9) had a significantly higher NAb titer for the prototype virus (β = 0.275, 95% CI = 0.298 ~ 3.339) (Supplementary Table S2). Compared with those who had never had COVID-19 (median log2 = 3), those who had had COVID-19 (median log2 = 5) had a significantly higher NAb titer for Omicron BA.5 (β = 0.222, 95% CI = 0.177 ~ 2.314) (Supplementary Table S3). Compared with nondrinkers (median log2 = 5), ever drinkers (median log2 = 4) and current drinkers (median log2 = 3) had significantly lower NAb titers for Omicron BF.7 (β = −0.225, 95%CI = −2.257~-0.026; β = −0.250, 95%CI = −2.951~−0.102) (Supplementary Table S4). Compared with nondrinkers (median log2 = 3), current drinkers (median log2 = 1) had a significantly lower NAb titer for Omicron BQ.1 (β = −0.274, 95%CI = −2.323~−0.183) (Supplementary Table S5).

Discussion

On May 5^th, 2023, 41 months after January 2020, the WHO declared that COVID-19 was now an established and ongoing health condition that no longer constituted a public health emergency of international concern (PHEIC), and control and prevention of the COVID-19 pandemic should transition to long-term management.³⁵ Therefore, close surveillance of SARS-CoV-2 evolution, immune evasion and vaccination are still essential for long-term management of COVID-19. In particular, understanding the immune status of residents will be the key for policy-makers to implement booster or nonpharmacological interventions. On January 8^th, 2023, China terminated the dynamic zero-COVID policy, lifted the requirement for quarantine upon entry³⁶ and was reopened based on mass vaccination (87.64% for primary vaccination, 40.32% for booster vaccination) and lower pathogenesis of Omicron VOCs. However, there is no information about the basic immunity status of Chinese residents with sole vaccination, complete vaccination or breakthrough infection. Returning travelers can serve as an example to address these issues.

In this study, we showed the basic immunity status of Chinese individuals with primary vaccination and homologous and heterologous booster vaccination through a survey of returning travelers before the reopening of China, which may mimic the circumstances of COVID-19 we will encounter in the near future. We first assessed the humoral immune responses of participants to SARS-CoV-2 by competitive ELISA. The results indicated that all participants had high concentrations of NAb. The participants in the heterologous booster group had a higher concentration of NAb than those in the primary and homologous booster groups. Although the participants enrolled in our study were all SARS-CoV-2 RNA negative, they may have recovered from infection before entering China. Surprisingly, many participants denied having a history of COVID-19 on the questionnaire, and this may have been influenced by the quarantine policy. A study by Pei Yu et al.³⁷ showed that Omicron variant breakthrough infection elicited higher specific memory immunity than a third-dose booster in healthy vaccinees. These findings were also in agreement with those of preliminary studies^38,39 that indicated that SARS-CoV-2 infection is a stimulus to induce enhanced immunity. Therefore, reinfection with SARS-CoV-2 may be a cause of higher NAb in participants without a booster. In our study, we also assessed the secretion of SARS-CoV-2 N and S proteins in the serum by ELISA. A few serum samples (19/108, 17.6%) were positive for N protein, which might be a sign of recent infection. Indeed, all 19 N protein-positive participants showed high NAb titers (184 GMTs against the prototype virus, ranging from 2-2048). This is related to the appearance and persistence of anti-N and anti-S markers after SARS COV-2 infection.⁴⁰ Studies have reported that anti-N protein antibodies persist longer than anti-S protein antibodies, with anti-N protein antibodies peaking at approximately 30 days postinfection.⁴¹

The protection effectiveness of COVID-19 vaccines decreases over time,^42,43 which raises questions on whether immunity from COVID-19 vaccination enables the prevention of breakthrough infection, severe illness or death. Particular concerns are related to the recently emerged highly transmissible Omicron VOCs.^44–46 We determined the NAb titers of 108 participants against the SARS-CoV-2 prototype virus and four Omicron VOCs (BA.5, BF.7, BQ.1 and XBB.1). We found a large proportion of NAb-positive participants and higher NAb titers against the prototype virus but significantly decreased titers against Omicron VOCs. For instance, the GMT of participants in the heterologous booster group dropped from 248 for the prototype virus to 40 for Omicron BA.5 (5.40-fold). A 6.76- and 31.00-fold decrease against Omicron BF.7 and BQ.1 or XBB.1 occurred, respectively. Although the NAb titers against Omicron VOCs generally declined with time, 52.9% (9/17), 35.3% (24/68) and 73.9% (17/23) of participants with primary vaccination, homologous booster vaccination and heterologous booster vaccination maintained neutralizing antibody levels against XBB.1, respectively, which may show a lower immune tolerance to XBB.1 in the future in China.

Cyclic exposure to the continuous generation of mutational variants may largely enhance herd immunity fortified by vaccination and natural infection.^47,48 Compared to the primary vaccination and homologous booster groups, higher NAb titers were found in the heterologous booster group, whether against the prototype virus or Omicron VOCs. These data indicated that vaccination with heterologous boosters has advantages in producing high levels of neutralizing antibodies.^49–51 Moreover, those who are vaccinated with the prototype virus vaccine had a high NAb titer against recently circulated Omicron VOCs, which may result from being either symptomatic or asymptomatic in the current epidemic wave. In this study, we found that 6 and 9 participants showed higher NAb titers to the BF.7 and BA.5 variants, respectively. This may indicate an either symptomatic or asymptomatic reinfection, although that was denied during the questionnaire survey. In addition, 6 participants showed the same NAb titers against BA.5 and BF.7 simultaneously. The epidemiology survey did not support coinfection of both viruses because of no spatiotemporal intersection for those participants in terms of geography.

Previous reports indicated that NAb titers decrease with time, whereas specific memory B and T cells can be maintained for 6 to 8 months in convalescent COVID-19 patients.^52–54 Here, we assessed MTC responses in 36 participants by detecting the production of IL-2 and IFN-γ, which are biomarkers of host immunity against intracellular pathogens, including SARS-CoV-2.⁵⁵ We found that SARS-CoV-2-specific IL-2- or IFN-γ-producing cells were variable in all participants. The proportions of positive samples in the primary vaccination group were comparable to those in the homologous and heterologous booster groups, which was inconsistent with archived studies. An investigation in Sweden indicated that heterologous vaccination (inactivated vaccine followed by an mRNA booster) stimulates a more significantly specific T-cell response.⁴⁹ This may have been influenced by the limited sample size in this study, particularly for the heterologous vaccination group. In addition, we found higher NAb titers in the heterologous vaccination group than in the other two groups. Nevertheless, the MTC response should be addressed in a large population study in the future.

The consistency of NAb titers with competitive ELISA NAb, MTC responses and N protein production of SARS-CoV-2 were also evaluated. NAb titers were positively correlated with competitive ELISA NAb. This finding was in line with that in one study that detected SARS-CoV-2 neutralizing antibody levels in COVID-19 patients and vaccines by automated chemiluminescent immunoassay (CLIA), and they showed a positive correlation between neutralizing antibodies and CLIA.⁵⁶ However, no obvious correlation was obtained between NAb titers and MTC responses, which was not consistent with prior reports that memory T-cell responses were correlated with neutralizing antibody responses.^57,58 This may be due to bias from the limited number of tested participants.

To assess the factors that may affect herd immunity, demographic and epidemiological information was evaluated with the decline in NAb titers in all 108 participants. We found that smoking, alcohol consumption, SARS-CoV-2 infection, COVID-19 vaccination, and the interval time between the last vaccination and sampling were independent influencing factors for NAb titers. Indeed, numerous studies have shown that COVID-19⁵⁹ and SARS-CoV-2 vaccination^60,61 are influencing factors for NAb titers. Overall, the immune response to the 3-dose vaccination schedule was significantly higher than that to the primary booster. We also found that smoking and alcohol consumption were associated with a higher risk of negative NAb titers, which agrees with the findings of previous studies.^62,63 However, the effect of alcohol consumption on NAb titers is quite controversial. Others have found that alcohol intake was not significant for NAb titers.⁶⁴ The present finding regarding alcohol usage may not be observed in populations with different genetic backgrounds, e.g., East Asians are genetically deficient in one of the enzymes involved in metabolizing alcohol (aldehyde dehydrogenase).⁶⁵ However, the influencing factors were variable for different VOCs. Among them, the vaccine dose was the unique common influencing factor for the four Omicron variants. A study also found vaccine type and the number of doses to be primary influencing factors for efficacy and immunogenicity.⁶⁶ Thus, vaccination strategies should be considered with high priority when challenged with the next novel VOCs.

In conclusion, for the first time, we reported the status of herd immunity among returning travelers to China by recruiting returning Chinese individuals in Guangzhou, China. We found broad and enhanced immunity of participants against prototype SARS-CoV-2, as well as a variety of Omicron VOCs. The comprehensive statistical analysis supported that vaccination was the key factor for host humoral immunity against SARS-CoV-2. Nevertheless, we should continue to focus on the long-term immunogenicity and effectiveness of COVID-19 vaccines in the future. The present data will extend our understanding of future complicated infection and vaccination circumstances and guide the long-term management of the COVID-19 pandemic in China, as well as other countries worldwide.

Author contributions

S.J., H.J., L.Y., L.B. and L.J. designed the study. Y.R., C.H., Y.L., Z.H., Z.P., L.C., L.H., Z.L., Z.X., R.Q. and H.X. all contributed to the experimental part of the study. L.X., C.Y., D.Y., L.Z., L.J., X.J. and X.X. contributed to data analysis. C.H. and L.X. drafted the manuscript, and S.J. and Y.R. revised the final manuscript. All authors contributed to data acquisition and data interpretation and reviewed and approved the final version.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary Material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2023.2300208.

Additional information

Funding

This work was supported by grants from the Guangdong Science and Technology Program [2022A1111090004, 2021B1212030007, 2021A0505110014].