PDF
Citation
Print
INTRODUCTION
The anti-nucleocapsid protein IgG antibody (N-IgG) response is increased by natural infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) [1,2]. N-IgG responses can be determined up to one year after the infection in 99% (109/110) of unvaccinated patients [3] and have an estimated half-life of 283 days in these patients [4]. N-IgG responses are not elicited by the current coronavirus disease 2019 (COVID-19) vaccines, which target the spike protein; therefore, N-IgG responses have been used to determine previous COVID-19 infections. However, a recent study reported that the seropositivity of the N-IgG response was 40% (21/52) in COVID-19 positive patients vaccinated with two doses of the mRNA-1273 vaccine [5]. This raised the question of the potential underestimation of seroprevalence with anti-N antibody testing in vaccinated populations.
Most of the world’s population has received at least one dose of COVID-19 vaccine [6]. In South Korea, 87% of the population has received two doses of COVID-19 vaccines [7]. Therefore, this issue of potential underestimation of seroprevalence is important for public health policy decision-making. However, whether vaccination status may affect the seropositivity of the N-IgG response in immunocompromised (IC) patients and non-immunocompromised (healthy) adults remains unclear. In this study, we evaluated the seropositive rate of N-IgG responses in healthcare workers (HCWs) and IC patients with PCR-confirmed COVID-19 who have received their third or fourth dose of COVID-19 vaccine [ChAdOx1 (AstraZeneca), BNT162b2 (Bio-NTech-Pfizer), or mRNA-1273 (Moderna)] and compared them with those in unvaccinated patients with COVID-19.
MATERIALS AND METHODS
1. Study population and sample collection
We enrolled HCWs who have been vaccinated against SARS-CoV-2 (either two doses of ChAdOx1 and a third dose of BNT162b2, three doses of BNT162b2, or three doses of mRNA-1273) between March and October 2021. We enrolled immunocompromised (IC) patients with liver or kidney transplants who have received at least two doses of homogeneous or heterogeneous vaccines (ChAdOx1, BNT162b2, and mRNA-1273) between June and December 2021. We also screened patients who had not received any doses of the COVID-19 vaccine and who had PCR-confirmed COVID-19 and were asymptomatic or had mild or moderate symptoms between March 2020 and November 2021. All the participants in this study were enrolled at the Asan Medical Center, a tertiary care hospital in Seoul, South Korea. To measure S1 and N-IgG responses, blood was sampled 2 weeks after every dose and then every 3 months from March 2021-July 2022 in the vaccinated HCWs and IC patients and 1 month after the COVID-19 diagnosis from April 2020-December 2021 in the unvaccinated patients. The timelines of the vaccinations and blood sampling are presented in Fig. 1A. This study was reviewed and approved by the Institutional Review Board of Asan Medical Center (IRB nos. 2020-0297 and 2021-0170). Written informed consent was obtained from all the participants.
The participants were divided into three groups according to COVID-19 infection status as follows: PCR-confirmed: patients with a PCR-confirmed COVID-19 diagnosis; infection-suspected case based on serologic results (infection-suspected): the participants suspected of hidden infection due to a two-fold or more increase in S1-IgG levels compared to those at least 3 months earlier; infection-naïve: the participants who were never confirmed with COVID-19 and showed waning of S1-IgG responses.
2. Antibody response measurement
Plasma was isolated from blood samples by centrifugation at 2500 rpm for 10 min and stored in a deep freezer at −80℃ until utilized. SARS-CoV-2 anti-spike 1 protein IgG antibody was quantified by an enzyme-linked immunosorbent assay (ELISA) that was designed in-house (details of which are described in our previous studies) [8,9]. Briefly, plasma was diluted at 1:100 through 1:100,000, added to 96-well plates (MaxiSorp; Thermo Fisher Scientific, Waltham, MA, USA) coated with the SARS-CoV-2 S1-His protein (2 μg/mL, SinoBiological, Beijing, China), and incubated for 2 h. Horseradish peroxidase-conjugated anti-human IgG (Jackson Immunoresearch, West Grove, PA, USA) was used as the detection antibody. The measured S1-IgG responses were presented as international units per milliliter (IU/mL), standardized with reference pooled sera from International Vaccine Institute (Seoul, South Korea).
We also measured the SARS-CoV-2 anti-nucleocapsid protein IgG antibody using a PCLOKII SARS-CoV-2 Dual IgG (PCL Inc., Seoul, Korea) approved by Korea Disease Control and Prevention Agency (KDCA), which is an automated fluorescence immunoassay based on ELISA [10].
3. Statistical analysis
Data are presented as means±standard deviations (SDs) for quantitative variables and frequencies and percentages (%) for qualitative variables. Categorical variables were compared using the chi-squared or Fisher’s exact tests. All tests of significance were two-tailed. Statistical significance was set at P≤0.05. The data were analyzed using SPSS version 24.0 (IBM Corp., Armonk, NY, USA).
RESULTS
1. Study participants
Among 59 hospitalized COVID-19 patients who were asymptomatic or had mild to moderate symptoms, 31 (including 1 asymptomatic, 18 with mild symptoms, and 12 with moderate symptoms) had no prior history of COVID-19 vaccination and agreed to participate in this study and blood sampling; therefore, these patients were included in the study. Of the 173 HCWs who had been vaccinated against COVID-19 vaccines, 61 spontaneously withdrew. Finally, 112 HCWs were enrolled in this study. Among them, 58 were vaccinated with two doses of ChAdOx1 and the third dose of BNT162b2 (ChA/ChA/BNT), 42 were vaccinated with three doses of BNT162b2, and 12 were vaccinated with three doses of mRNA-1273. Among IC patients (89 patients with liver transplant [LT] and 27 with kidney transplant [KT]), 59 patients with LT and 18 patients with KT who received at least two doses of COVID-19 vaccine and participated in this study until 9 months after the first dose were included. Participants were divided into three groups according to their COVID-19 infection status, as explained earlier. The study flowchart is presented in Fig. 1B, and the demographic characteristics are presented in Table 1.
2. Antibody responses
In unvaccinated PCR-confirmed patients, antibody responses were measured 1 month after the diagnosis (median 34 days, 95% confidence interval [CI], 29-39 days), and the positive rate of N-IgG responses was 93.5% (29/31) (Table 2). In the vaccinated HCWs, antibody responses were determined 6 months after the third vaccine dose, with a median of 51 days after the diagnosis (95% CI, 37-72 days) in PCR-confirmed HCWs. The median time between the third dose and the diagnosis was 124 days (95% CI, 115-139 days). The positive rate of N-IgG responses was 0% (0/51), 75.0% (39/52), and 66.7% (6/9) in infection-naïve, PCR-confirmed, and infection-suspected HCWs, respectively. In IC patients, antibody responses were determined at 6 months after the third dose in patients with LT and 3 months after the third dose in patients with KT, with a median of 48 days after the diagnosis (95% CI, 40-108 days) in PCR-confirmed IC patients. Eight of the 15 IC patients (PCR-confirmed) received only two doses of vaccines. The antibody responses in these participants were determined using the first blood sample taken at least 14 days after the diagnosis. The median time between the second or third dose and the diagnosis was 106 days (95% CI, 69-120 days). The positive rates of S1-IgG responses were 83.7% (41/49), 93.3% (14/15), and 100% (13/13). Moreover, the positive rates of N-IgG responses were 0% (0/49), 60.0% (9/15), and 61.5% (8/13) in infection-naïve, PCR-confirmed, and infection-suspected IC patients, respectively. The positivity rate of N-IgG responses in unvaccinated PCR-confirmed patients (93.5%) was significantly higher than that in vaccinated PCR-confirmed HCWs (75.0%) (P=0.041, Table 2). The difference in the positivity rate of N-IgG responses between vaccinated PCR-confirmed HCWs and IC patients was not significant (P=0.332).
Furthermore, the IC patients (PCR-confirmed) were classified into two subgroups (LT and KT). The positive rate of N-IgG responses in patients with LT was higher (70%, 7/10) than that in patients with KT (40%, 2/5); however, the difference was not significant (P=0.329, Supplementary Table 1). Among the 15 PCR-confirmed IC patients, 13 received two or three doses of BNT162b2, and the positive rate of N-IgG response was 53.8% (7/13). The remaining two patients were vaccinated with ChAdOx1 and mRNA-1273 each and showed positivity for N-IgG responses. We also performed subgroup analyses for the positivity of N-IgG responses in vaccinated HCWs depending on the types of vaccines. In the PCR-confirmed HCWs, the positive rates of N-IgG responses were (24/30) 80.0%, (14/19) 73.7%, and (1/3) 33.3% in the ChA/ChA/BNT, BNT162b2, and mRNA-1273 groups, respectively. The N-IgG responses in the mRNA-1273 groups were significantly lower than those in the ChA/ChA/BNT groups (P=0.048). The detailed data are shown in Supplementary Table 2.
DISCUSSION
The results of this study showed that the positivity of N-IgG responses was lower in vaccinated (75.0%) than in unvaccinated participants (93.5%) among the patients with PCR-confirmed COVID-19. Our previous study reported that anti-spike 1 IgG antibody responses were higher in patients with a severe or critical infection compared with those of patients with a mild infection [8]. The occurrence of higher N-IgG responses was also found in severely or critically infected patients [11]. COVID-19 vaccines prevent the development of severe diseases, and vaccines including ChAdOx1, BNT162b2, and mRNA-1273 demonstrated an estimated 70% effectiveness against severe disease [12]. Similarly, the KDCA estimated that COVID-19 vaccines have an 80% preventive effectiveness against severe disease or death [13]. Severe infection was associated with higher levels of antibody responses [14]; therefore, the lower positivity of N-IgG responses in vaccinated participants could be partly explained by the vaccine’s effectiveness against severe disease, which has been associated with lowered levels of N-IgG responses.
The currently available COVID-19 vaccines all target the spike protein. Therefore, the vaccines elicit anti-spike protein antibody responses and develop a memory response against the spike protein [15]. Our data showed that the mean S1-IgG response was five times higher in the PCR-confirmed HCWs (5,423±3,490 IU/mL) compared with the infection-naïve HCW (1,065±1,348 IU/mL) among the vaccinated HCWs. When COVID-19 infection occurs in vaccinated persons, the nucleocapsid protein antibody responses may be attenuated by the greater effect of the spike protein memory responses against SARS-CoV-2. However, the exact mechanism contributing to the lowering of N-IgG responses in vaccinated individuals who become infected with SARS-CoV-2 remains unknown.
The seroconversion of N protein antibodies was lower in vaccinated populations compared with that of unvaccinated populations diagnosed with COVID-19 [5,16]. The positive rate of N-IgG responses in patients with PCR-confirmed COVID-19 was 26% in those who received the BNT162b2 vaccine and 40% in those who received the mRNA-1273 vaccine [5,16]. Our data showed 75% positivity for N-IgG responses in HCWs with PCR-confirmed COVID-19 vaccinated with ChAdOx1, BNT162b2, or mRNA-1273. This result is higher than the seropositivity rates of N-IgG responses in previous studies. Furthermore, in previous studies, the median time between the second dose and the COVID-19 infection was 30 days in the BNT162b2 recipients and 77 days in the mRNA-1273 recipients, whereas the median time between the third dose and the COVID-19 infection was 124 days in our study. Spike antibody responses induced by COVID-19 vaccines peak approximately 1 month after vaccination and wane over time [9,17]. At 6 months after the third dose, the S1-IgG responses in infection-naïve HCWs were 700±490, 765±1021, and 3,096±2,158 IU/mL and the positive rates of N-IgG responses in PCR-confirmed HCWs were 80.0%, 73.7%, and 33.3% in the ChA/ChA/BNT, BNT162b2, and mRNA-1273 groups, respectively (Supplementary Table 2). It is believed that the seroconversion of N-IgG may be lowered as spike antibody responses induced by vaccines increase. Therefore, the different effects of vaccination on the seropositivity of N-IgG responses between studies may be partially caused by the heterogeneity of vaccine-induced waning immunity. Taken together, these data provide important insights into the interpretation of seroprevalence data based on N-IgG responses against SARS-CoV-2 infection in highly vaccinated populations.
This study had a few limitations. First, our sample size was small, particularly for the subgroup analyses (only 12 participants received the mRNA-1273). However, we included the most common types of COVID-19 vaccines used in the Korean population and unified the schedules of follow-up and blood sampling. Therefore, our findings could represent the real seroprevalence. Second, we only used the automated fluorescence immunoassay based on ELISA (PCL Inc.) for the detection of N-IgG, which has a 98.9% sensitivity and a 99.3% specificity using samples collected 14 or more days after symptom onset [18]. The results might have differed if the N-IgG responses were measured using assays from different manufacturers, which use other epitopes or methodologies. Third, the COVID-19 strains were different in some of our participants. Unvaccinated patients were infected with the original strain and Delta variant, but vaccinated participants were infected with the Omicron variant. The positive rate of N-IgG responses in vaccinated individuals infected with the original strain is unknown since COVID-19 vaccines were not yet developed at that time. However, the anti-nucleocapsid IgG antibody responses were similar among the 18 patients infected with the Delta variant and the 51 infected with the Omicron variant [19]. Future studies are warranted to determine the differences in the seropositivity of N-IgG responses among different variants of the COVID-19 infection.
In conclusion, vaccinated patients showed lower seropositivity of N-IgG responses compared with those of unvaccinated patients after COVID-19 infection. Our findings suggest that vaccination affects seroconversion of the anti-nucleocapsid protein antibody. Therefore, an underestimation of the seroprevalence of COVID-19 infection based on N-IgG responses may be possible in vaccinated populations.
SUPPLEMENTARY MATERIALS
Supplementary materials can be found via https://doi.org/10.14192/kjicp.2023.28.1.92.
XML Download