Introduction

Coronavirus Disease 2019 (COVID‐19) is caused by Severe Acute Respiratory Syndrome Coronavirus-2 (SARS‐CoV‐2) virus. It causes clinical symptoms ranging from asymptomatic to severe respiratory failure even in the era of new variants [1]. The COVID-19 pandemic continues to spread globally with emergence of new variants which have the ability to spread more rapidly than the previous variants [2]. When compared to the Delta variant, the Omicron variant BA.1/BA.2 is associated with reduced clinical severity and hospitalization risk in both children and adults [3]. Adults have significant Omicron-specific humoral responses, but data on children is limited. We previously reported that half of children infected with the Wuhan strain seroconverted, compared to adults [4].

The mortality rate of COVID-19 infection in children is less than in adults [5]. After the emergence of the Omicron variant, increased rate of SARS-CoV-2 infection was reported in children [6]. Children have lower hospitalization rates for COVID-19 than adults but serious illness and death have occurred [7]. Children hospitalized with COVID-19 were admitted to ICU in 28% to 40% of cases, invasive mechanical ventilation was necessary in 6% to 18% of cases, and up to 3% died [8,9]. SARS-CoV-2 infection in children can lead to significant consequences such as acute respiratory distress syndrome, myocarditis, and multisystem inflammatory syndrome in children (MIS-C) [8]. MIS-C was diagnosed based on the presence of antibody in individuals suspected of having obtained the virus in the community or through household contact, or on the basis of a prior history of COVID-19 infection [10].

After the initial infection, the development of antibodies is a critical factor in comprehending the dynamics of COVID-19 infection [11]. Antibodies are essential components of the immune system that can neutralize pathogens and prevent their entry into healthy cells. However, existing studies have shown conflicting results regarding the levels of antibodies and durability in non-vaccinated who experience a SAR-CoV-2 infection [12–15]. Understanding how COVID-19 antibodies work in non-vaccinated children with previous COVID-19 infections is necessary to inform public health policies and guidelines. Developing a clear understanding of how long children’s immune systems remain immune to the virus after the initial infection is important when considering the pandemic’s trajectory in the months and years ahead. Therefore, a more comprehensive assessment of SARS-CoV-2 antibody dynamics in non-vaccinated children with previous COVID-19 infections is essential. Such information may help to provide a better understanding of how COVID-19 antibodies work and how the immune system responds to the virus in children. Furthermore, it can facilitate the development of more effective public health policies and interventions to protect children and prevent the spread of COVID-19.

Due to lack of data on SARS-CoV-2 antibodies in children following Delta or Omicron infection, particularly in Asian populations, this study was conducted to determine antibodies against SARS-CoV-2 in children hospitalized due to COVID-19 during the era of pre-Omicron and Omicron variants.

Materials and methods

Results

Studied population and characteristics

There were 119 hospitalized children due to COVID-19 included in our study, of whom 59.66% (n = 71) were male. The median age and interquartile range (IQR) of children was 49 months (17–87). All of them were confirmed COVID-19 by rtPCR. Twenty-four children (20.17%) had at least one co-morbidity. Asthma and allergic rhinitis were the most common accounted for 7 and 6 children, respectively. Eight asymptomatic patients were admitted with a high-risk condition (asthma, congenital heart disease, morbid obesity) who required close monitoring. A total of 83 cases (69.74%) had pulmonary infiltration from chest radiograph. They were diagnosed as COVID-19 pneumonia and classified as moderate to severe COVID-19. Only 2 of them had critically ill severity. Most of our patients received favipiravir as an antiviral therapy. All of the cases were unvaccinated against COVID. Fifty-eight children were infected during pre-Omicron, and 61 children were infected during Omicron era. The gender, comorbidities, chest radiograph classification were not statistically different. The median age of children during Omicron were significantly lower than pre-Omicron era (69 and 38 months, p = 0.009) (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Baseline characteristics of children infected with COVID-19 during pre-Omicron and Omicron era.

https://doi.org/10.1371/journal.pone.0297991.t001

COVID-19 infection in Omicron era is associated with lower platelet count but higher C-reactive protein level compared with pre-Omicron

The blood investigation results during hospitalization were presented in Table 2. The hemoglobin, and total white blood cell count were similar in both groups. The median platelet count in children infected during pre-Omicron era was higher than in children infected during the Omicron era (324,000 and 283,000, p = 0.008). The median level of C-reactive protein (CRP) in children infected during Omicron era was significantly higher than the median level of CRP during pre-Omicron era (p = 0.023).

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Serum levels of complete blood count, C-reactive protein, and COVID-19 antibody at diagnosis in children infected with COVID-19 during pre-Omicron and Omicron era.

https://doi.org/10.1371/journal.pone.0297991.t002

COVID-19 infection in Omicron era associated with lower level of SARS-CoV-2 antibody compared with pre-Omicron

The median (IQR) of SARS-CoV-2 antibody levels at the time of diagnosis was 0.40 (0.40–3.00) U/mL during pre-Omicron and 0.40 (0.40–0.40) U/mL during Omicron. The overall median time (IQR) of follow-up was 79 days (72–95 days), which divided into the 77.5 days (72–86) and 88 days (74–104) after diagnosis, for children infected during pre-Omicron and Omicron era, respectively. At follow-up time, children infected during pre-Omicron and Omicron remained seropositive with the median of anti-SARS-CoV-2 antibody level of 206.1 (87.9–424.1) U/mL. The median level of Anti-SARS-CoV-2 antibody at follow-up of the children infected during pre-Omicron was 350.20 (IQR: 111.50–661.20) U/mL and that of children infected during Omicron was 165.20 (IQR 78.20–284.80) U/mL (p = 0.004) as shown in Fig 1. Children infected during pre-Omicron had mean of Anti-SARS-CoV-2 antibody levels at follow-up significantly higher than children infected during Omicron [mean difference (MD) = 292.57, 95% CI: 53.85–531.28, p = 0.017). No difference in mean of Anti-SARS-CoV-2 levels was found when the study participants were compared based on gender, age, co-morbidities, chest radiograph classification, and diagnosis (Table 3). The regression model adjusting for the covariates confirmed a significant difference (p = 0.017) in the Anti-SARS-CoV-2 antibody levels between children infected during pre-Omicron and Omicron.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Anti-SARS-CoV-2 levels at diagnosis and follow-up in children infected with COVID-19 during pre-Omicron and Omicron era.

https://doi.org/10.1371/journal.pone.0297991.g001

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Factors associated with SARS-CoV-2 antibody level of children infected with COVID-19 at follow-up.

https://doi.org/10.1371/journal.pone.0297991.t003

Discussion

In this observational study, the level of SARS-CoV-2 antibodies in unvaccinated hospitalized children due to COVID-19 during pre-Omicron and Omicron era. Here, we found that children infected during Omicron era had lower SARS-CoV-2 antibody levels compared to pre-Omicron era. A previous study suggested that Omicron SARS-CoV-2 virus has lower virulence and lower activation of the host immune response [20,21]. Alternatively, the strains in pre-Omicron, especially the Delta variant showed more severe symptom and higher immune activation [22,23]. In our study, we found that the antibody levels at diagnosis during pre-Omicron was slightly greater than Omicron. The potential for confounding the antibody levels determined using the Elecsys assay exists due to the incongruity between the antigen utilized to quantify RBD antibodies and the infecting Omicron variants. As stated, the double-antigen principle of the Elycsys assay introduces a bias towards high-affinity antibody binding results when employing the receptor binding domain of the index strain. The pre-omicron era was characterized by the preponderance of the Beta and Delta variants, which each had a mere three and two RBD mutations, respectively. On the contrary, the preponderant Omicron BA.1, BA.2, and BA.5 variants exhibited a total of thirteen, sixteen, and seventeen RBD mutations during the studied period [24]. Consequently, antibodies generated against the pathogenic Omicron variant may potentially encounter numerous mismatches and fail to bind to the index spike RBD protein with significant affinity. So, the inquiry arises as to whether the antibody levels induced by Omicron are diminished, or whether the assay’s measurement of diminished binding is the result of discrepancies with the antigen utilized in the assay. At follow up, children infected during pre-Omicron had SARS-CoV-2 antibody levels higher than children infected during Omicron. This finding indicates that the increased immune activation was also present at the follow-up, highlighting that immune activation in Omicron is lower than pre-Omicron.

In our study, we also found that the presence of SARS-CoV-2 antibodies was not associated with the severity of illness. This result contradicts a prior study conducted on adults during the era of the delta variant, which demonstrated that elevated levels of SARS-CoV-2 neutralizing antibodies were associated with greater severity [25]. A study in adult showed that more severe COVID-19 infections resulted in a more robust humoral immune response, this finding may indicate that the mild illness in the population of children may not produce strong immune response [26]. It is also important to note that the presence of SARS-CoV-2 antibodies does not necessarily correlate with protection against future infections, several studies have shown that T-cell responses and memory B-cells may also play a role in immune protection against COVID-19 [27–32].

Several studies compared the SARS-CoV-2 neutralizing antibodies before and after vaccination [33,34]. Few studies investigated the unvaccinated children population. Comparing the initial diagnosis and post-infection levels of SARS-CoV-2 neutralizing antibodies in both pre-omicron and omicron era, we discovered that their presence is extremely low in our study. In contrast to the results in previous study of post vaccinated SARS-CoV-2 neutralizing antibodies in all kind of vaccines, they showed the higher level of SARS-CoV-2 antibodies [35]. A previous study has shown that COVID-19 neutralizing antibodies are generally lower in children following a natural infection with SARS-CoV-2 compared to vaccination with mRNA COVID-19 vaccines. This may be due to differences in the immune response between natural infection and vaccination [15]. This difference in the level of SARS-CoV-2 antibodies between natural infection and vaccination may be due to the composition and concentration of the antigen in the vaccine which is specifically designed to stimulate a strong and targeted immune response [36]. In contrast, natural infection with SARS-CoV-2 can result in a less targeted and weaker response, which may explain the lower levels of COVID-19 antibodies [13]. Despite the development of SARS-CoV-2 antibodies in unvaccinated children who have had COVID-19 infections, spontaneous infection in children may not provide adequate immunity to prevent subsequent infections, and vaccination remains the primary preventative approach [37]. This finding highlights the need for vaccination even, after natural infection as the immune response from natural infection has been shown not to be robust [38].

COVID-19 antibodies test can provide several benefits to children who have been exposed to or infected with SARS-CoV-2. Measuring COVID-19 antibodies in children can help determine whether they have been exposed to the virus, even if they were asymptomatic or had mild symptoms [39]. This is important for identifying infected individuals, tracking the spread of the virus, and making decisions about public health measures like quarantine and contact tracing [40]. COVID-19 antibodies test could be one key to identify MIS-C. This is a rare but severe condition that has been observed in some children who tested positive for COVID-19. One way to identify MIS-C is by testing for COVID-19 antibodies in unvaccinated children who have been affected by the virus [41]. Children who develop MIS-C typically have high levels of COVID-19 antibodies, which indicate that their immune system has been activated to fight the virus. Therefore, measuring COVID-19 antibodies could be a valuable tool in identifying children at risk for developing MIS-C [42]. Testing for COVID-19 antibodies in children with suspected MIS-C can improve the diagnostic accuracy and provide an optimal time for early medical intervention.

Our findings also demonstrate a statistical difference in CRP and platelet count in Omicron group vs pre-Omicron group. However, the levels of CRP and platelet were still within the normal range. Therefore, this difference appears not to be clinically significant. We also discovered a difference in the age group of Omicron variant that is lower than delta, which could suggest that the Omicron variants are more contagious [43].

Our study has some limitations that should be mentioned. For the recruited participants, the subjects were mild to moderate cases, no participant had severe illness hence it can be found less antibodies response in our population. Although our study is a retrospective study, standardized measurement protocols were employed in our study to ensure the collection of comprehensive data with minimal missing values. Furthermore, the strain type of virus of the participants are unknown. Age of the subjects between groups was significantly different. This factor might affect the immune status and the SARS-CoV-2 antibody level. Future study with multicenter approach to enroll more population with detailed viral strains is suggested.

Conclusion

Our study showed that the host immune response in Omicron is different compared to pre-Omicron era by showing lower the SARS-CoV-2 antibody level. The Omicron seems to infect young children. Children do develop low level of SARS-CoV-2 antibodies in response to natural infection with SARS-CoV-2. These findings suggest that vaccine prioritization should be directed towards unvaccinated children. The immunization strategies and policies for children, particularly in areas where vaccine access is limited should be implemented worldwide.

Acknowledgments

We would like to thank staff of the Research and Diagnostic Center for Emerging Infectious Diseases (RCEID) for their helpful discussions of the manuscript.