Seroepidemiological assessment of SARS-CoV-2 vaccine responsiveness and associated factors in the vaccinated … – Nature.com

Basic characteristics of the study participants and overall seroprevalence among the vaccinated community

The distribution of participants based on socio-demographic data and potential risk factors for SARS-CoV-2 infection is presented in Table 1. Out of the total, 10,669 blood samples were assessed (12% of participants refused to participate in the serological survey), comprising 7,380 women and 3,475 men, with a median age of 44years (ranging from 14 to 102years) (Table 1).

The serological survey encompassed 10,669 participants, with 8,602 having received two doses of BBIBP-CorV/Sinopharm (meanSEM of months after the second dose=8.110.08), 1,817 with ChAdOx1 nCoV-19 (COVISHIELD, Oxford/AstraZeneca; meanSEM of months after the second dose=9.820.15), 208 with BNT162b2 (Pfizer-BioNTech, Comirnaty, Pfizer, USA; meanSEM of months after the second dose=7.140.55), and 42 participants vaccinated with a single dose of JCovden/Johnson & Johnson's COVID-19 vaccine (meanSEM of months after the first dose=7.501.00). Adjusted overall seroprevalences did not show significant differences between vaccines (p=0.099).

We determined an adjusted seroprevalence rate of 96% (95% CI: 95.5% to 96.4%) among participants who received the BBIBP-CorV vaccine (Table 2). Logistic regression analysis revealed several variables associated with a higher risk of detecting anti-SRAS-CoV-2 antibodies (Table 3). Specifically, a higher prevalence of antibody positivity was linked to age. Additionally, the odds of being antibody-positive are 0.59 times lower for male participants than for females. A higher prevalence of antibody positivity was associated with hypertension, mask-wearing, and PCR-confirmed COVID-19 disease (Table 3). Furthermore, the analysis by chronic kidney disease revealed that participants with kidney disease had a significantly lower probability of being seropositive than participants without kidney disease (Table 3).

For participants vaccinated with ChAdOx1-nCov-19, the adjusted seroprevalence was 97% (95% CI: 9697.9) (Table 2). Logistic regression analysis assessing the association between SARS-CoV-2 seropositivity and demographic characteristics and chronic diseases between February and June 2022 is presented in Table 4. In univariable analysis, participants in the age groups of more than 1419years and those with a PCR-confirmed diagnosis (OR=2.65; 95% CI: 1.056.68; p=0.038) tended to be significantly more likely to be seropositive for anti-SARS-CoV-2 antibodies (Table 4). In contrast, in the multivariable analysis, only participants with a PCR-confirmed diagnosis were more likely to be seropositive for anti-RBD (OR=2.69; 95% CI: 1.177.78; p=0.036) (Table 4).

For participants vaccinated with BNT162b2, the overall adjusted prevalence of anti-SRAS-CoV-2 antibodies was 98.5% (95% CI: 95.0100.0) (Table 2). Logistic regression analysis showed that participants with hypertension who received BNT162b2 were more likely to be seronegative compared to female participants and those without hypertension (Table 5).

The adjusted seroprevalence among participants vaccinated with the Janssen/Johnson & Johnson's COVID-19 vaccine was 98% (95% CI: 85.2100.0). Owing to the small sample size (n=42), logistic regression analysis of the association between SARS-CoV-2 seropositivity and demographic characteristics and comorbid conditions was not conducted.

Subsequently, we evaluated the magnitude of the humoral response by measuring IgG antibodies to the RBD of the S1 subunit of the SARS-CoV-2 spike protein. The median RBD antibody concentrations were 2355 AU/mL, 3714 AU/mL, 5838 AU/mL, and 2495 AU/mL after two doses of BBIBP-CorV/Sinopharm, ChAdOx1 nCoV-19/Oxford/AstraZeneca, BNT162b2/Pfizer-BioNTech, and after one dose of JCovden/Johnson & Johnson's COVID-19 vaccine. Significant differences were observed among vaccine brands (p<0.0001). Notably, there was no significant difference between the JCovden vaccine and the BBIBP-CorV vaccine (p=0.691) (Fig.1).

Anti-SARS-CoV-2 antibody levels among four vaccine brands. Data are presented as median and interquartile range for IgG antibody. MannWhitney and KruskalWallis tests were used.

Stratifying participants who received the BBIBP-CorV vaccine revealed a significant difference in antibody concentration titers by gender (median=2428 vs. 2238 AU/mL for females and males, respectively) (p=0.004) (Fig.2A). An association was identified between age and anti-RBD IgG levels in BBIBP-CorV vaccine recipients (<0.0001), with the highest levels observed in those aged65years (median=5145.5 AU/mL) (Fig.2B).

Anti-RBD IgG antibody responses to BBIBP-CorV vaccine/Sinopharm in the general population. (A) Antibody levels subdivided by gender. (B) Anti-SARS-CoV-2 IgG levels by age. (C) SARS-CoV-2 antibody titer in participants with and without diabetes. (D) SARS-CoV-2 antibody titer in participants with and without chronic kidney disease. (E) SARS-CoV-2 antibody titer in participants with and without hypertension. (F) SARS-CoV-2 antibody titer in participants with and without cancer. (G) SARS-CoV-2 antibody titer in participants with and without mask-wearing. (H) Antibody levels by the history of coronavirus disease 2019 (COVID-19). Data are presented as median and interquartile range for IgG antibody titers. MannWhitney test was used for comparisons.

Unexpectedly, participants with comorbidities exhibited the highest levels of anti-RBD IgG (Fig.2CE), while for participants with cancer, antibody levels showed no significant difference between those with (median=2078 AU/mL) and without cancer (median=2358 AU/mL) (Fig.2F). Additionally, individuals reporting mask-wearing demonstrated higher anti-RBD antibody titers (median=2468 AU/mL) compared to those not wearing masks (median=2215 AU/mL) (p=0.0002) (Fig.2G).

Lastly, our data revealed that individuals with confirmed exposure to SARS-CoV-2 had elevated anti-RBD antibody titers (median=3019 AU/mL) compared with uninfected individuals (median=2222 AU/mL) (Fig.2H).

Stratifying participants who received the Covishield vaccine revealed no significant difference in the humoral response by gender (median=3698 vs. 3845 AU/mL for females and males, respectively) (p=0.681) (Fig.3A). In contrast, bivariate Spearman analysis revealed a positive correlation between age and anti-RBD IgG titers (r=0.240, 95% CI: 0.195 to 0.284, p<0.0001) (Fig.3B).

Antibody responses against RBD after two doses of ChAdOx1-nCoV-19/AstraZeneca. (A) Gender difference in antibody response. (B) Scatter plot of the distribution of antibody titers according to age. (C) SARS-CoV-2 antibody titer in participants with and without diabetes. (D) SARS-CoV-2 antibody titer in participants with and without chronic kidney disease. (E) SARS-CoV-2 antibody titer in participants with and without hypertension. (F) SARS-CoV-2 antibody titer in participants with and without cardiovascular disease. (G) SARS-CoV-2 antibody titer in participants with and without cancer. (H) SARS-CoV-2 antibody titer in participants with and without mask-wearing. (I) Antibody levels by the history of coronavirus disease 2019. Data are presented as median and interquartile range for IgG antibody titers. Spearman correlation and MannWhitney test were used for comparisons.

Comparison between participants according to medical comorbidities revealed elevated anti-RBD antibody concentrations in those with diabetes (median=4660.5 vs. 3461 AU/mL for with diabetes and without diabetes, respectively) (p=0.002) (Fig.3C), chronic hypertension (median=4416 vs. 3546 AU/mL for with chronic hypertension and without chronic hypertension, respectively) (p=0.015) (Fig.3D), and renal disease (median=10,933 vs. 3637 AU/mL for with renal disease and without renal disease, respectively) (p<0.0001) (Fig.3E). In contrast, there was no statistical difference in antibody titers between participants with cancer (median=6038 AU/mL) and those without cancer (median=3708.5 AU/mL) (p=0.525) (Fig.3F) and cardiovascular disease (median=4094 vs. 3701 AU/mL for with cardiovascular disease and without cardiovascular disease, respectively) (p=0.830) (Fig.3G). Additionally, wearing a mask did not affect the humoral response in Covishield-vaccinated participants (Fig.3H) (p=0.611). However, prior exposure to COVID-19 increased the level of anti-RBD IgG (median=4618 vs. 3508 AU/mL for individuals with confirmed exposure to SARS-CoV-2 and uninfected individuals, respectively) (p=0.003) (Fig.3I).

In fully vaccinated participants with BNT162b2/Pfizer, stratification by demographics, comorbidities, and history of COVID-19 showed no significant differences in anti-RBD antibody concentrations (Fig.4AD,F). In contrast, an elevated antibody level was observed in participants who reported wearing a mask (median=6753 AU/mL) compared with those who did not report wearing a mask (median=4909 AU/mL) (p=0.046) (Fig.4E).

Antibody responses against RBD following two doses of BioNTech162b2/Pfizer vaccine. (A) Distribution of antibody titers according to sex. (B) Correlation of age and anti-RBD IgG antibody levels. (C) SARS-CoV-2 antibody titer in participants with and without diabetes. (D) SARS-CoV-2 antibody titer in participants with and without hypertension. (E) SARS-CoV-2 antibody titer in participants with and without mask-wearing. (F) Antibody levels by the history of coronavirus disease 2019. Data are presented as box and whisker plots with the minimum and maximum range for IgG antibody titers. Spearman correlation and MannWhitney test were used for comparisons.

Stratification of participants vaccinated with JCovden/Johnson & Johnson's COVID-19 vaccine showed no association by gender (p=0.456), and no correlation between age and anti-RBD antibody levels was noted (p=0.362). However, stratification of participants by comorbidities was not performed due to the limited sample size (n=42).

Read more:

Seroepidemiological assessment of SARS-CoV-2 vaccine responsiveness and associated factors in the vaccinated ... - Nature.com