Although the presence of neutralizing antibodies against SARS-CoV-2 proteins does not fully protect against infection, it significantly reduces its risk. Letizia et al. proved that seropositive convalescent young adults had as much as five times lower risk of reinfection compared to a seronegative comparison group [8]. Hence, although seroconversion is not a guarantee of effective neutralization activity, it does indicate the acquisition of at least partial immunity [9]. Moreover, the early adaptive humoral immune response involving seroconversion has a proven effect on slowing down the replication of the virus during infections and its faster elimination [10]. In addition, slow elimination and persistence of the virus can lead to complications such as mutations in viral proteins and long-COVID symptoms [11]. Immunocompromised children still have low vaccination rates. Additionally, they are exposed to significant clinical implications of SARS-CoV-2 infections, such as delay and modification in the treatment of their underlying disease [3], deterioration of graft function [4], and a tendency for virus persistence predisposing to the development of long-COVID symptoms, as well as reactivation of latent herpesviruses [11]. Therefore, the question of whether this group can develop an effective immune response to reinfections and how long it may persist is incredibly important.

In research studies regarding the presence of antibodies against SARS-CoV-2 among immunocompromised children, the authors have focused on the seroprevalence in specific populations rather than on the seroconversion of children with confirmed SARS-CoV-2 infections. Therefore, we aimed to explore the latter issue. Mayanskiy et al. revealed a seroconversion of 92% in 18 pediatric oncology patients by 3 weeks post-infection and in 100% by 6 weeks post-infection, which is similar to our results. By 18 weeks after the onset of COVID-19, this seropositivity rate declined to 54%. However, the authors did not compare the level of antibodies with an immunocompetent control group [12]. On the other hand, researchers from Hungary assessed the level of antibodies in 10 children with cancer at an interval of 1 to 4.5 months after the onset of COVID-19. They detected a seroconversion level of only 60% [13].

As for pediatric solid organ transplant (SOT) recipients, Talgam-Horshi et al. revealed a 96% seroconversion rate among 25 patients. However, they did not present the antibody levels or compare them with a control group [14]. Meanwhile, researchers from Saudi Arabia observed eight pediatric kidney transplant (PKT) recipients and showed their ability to maintain detectable levels of antibodies at a median follow-up time of 75 days [15]. We, in turn, observed these patients to have a detectable antibody titer for at least 90 days. None of the above-mentioned authors reported the number of patients reinfected with SARS-CoV-2 during their observation, despite having detectable titers of antibodies against the S protein. In our study, the reinfection rate was 5.5% (n = 3; one immunocompetent and two immunocompromised patients).

Various researchers have also assessed lymphocyte subsets in immunocompetent children with COVID-19. They proved that children, unlike adults, rarely have lymphopenia in the course of SARS-CoV-2 infections, and if present, its degree correlates with the severity of the disease and usually affects all lymphocyte subsets [16]. Our study did not reveal a tendency for lymphopenia in children, regardless of the severity of COVID-19. However, as for the values of individual lymphocyte subsets, the scientific data is not unequivocal. According to Mahmoudi et al., children with SARS-CoV-2-induced pneumonia have higher CD8 + T cell counts and lower CD4 + /CD8 + T cell ratios [15]. On the one hand, CD8 + T cells are responsible for destroying virus-infected cells and thus limiting their replication and spread [17]. On the other hand, they can cause damage to lung tissue if they are not sufficiently regulated [18]. However, researchers from China present differing results, according to which the more severe the course of COVID-19 and greater lung involvement, the higher the CD4 + /CD8 + T cell ratio [19]. This is in line with our results and could at least partially explain the milder course of SARS-CoV-2 infections, usually asymptomatic or mild without lung involvement, in immunocompromised children with significantly lower CD4 + /CD8 + ratios. It should be noted, however, that our study population differed from the groups mentioned above, which may substantially impact the differences in obtained results.

One of the most important features we observed in the lymphocyte subsets of immunocompetent children was the reduction of NK cells during COVID-19, which may be due to their participation in the defense against SARS-CoV-2. None of the above-mentioned studies in children has addressed this issue. However, it has been studied in more detail among adults. NK cells are part of innate immunity and play a key role in tackling viral infections. They eliminate virus-infected cells either directly through the degranulation of cytotoxic granules or through the secretion of cytokines and chemokines that modulate the activity of other immune cells [20]. Many researchers have demonstrated a reduced number of NK cells in patients with COVID-19 [21,22,23]. Krämer et al. proved that the activity of NK cells producing IFN-γ and TNF-α in response to SARS-CoV-2 infection leads to a reduction of viral proteins [21]. Unfortunately, these cells have also been shown to be dysfunctional in severe COVID-19, which reduces their participation in antiviral immunity [21, 24]. To the best of our knowledge, this phenomenon has not been analyzed in children, who are often characterized by different immune mechanisms. Thus, we are the first to report a reduced number of NK cells in the immunocompetent pediatric group during COVID-19, but unfortunately, we did not examine the activity of these cells. Therefore, this issue requires further research.

Our study has some limitations. The first one is a small size of the study group, especially the subgroup in which lymphocyte subsets were assessed. The second one is the disparity between the course of SARS-CoV-2 infection in immunocompetent and immunocompromised children, which can make it difficult to deduct conclusions on the protective effect of a lower CD4/CD8 ratio. However, due to the generally milder course of COVID-19 in the immunocompromised group of children, this limitation is difficult to omit. Another limitation is the lack of evaluation of a specific T-cell response to the infection in both groups of children, which would significantly complement the information about the humoral response. Although the specific T-cell response is the most important element of antiviral immunity, yet impossible to perform in our Institute at the beginning of our study, we decided to base on information about the humoral response, especially since antibody titers have been proven to correlate with the number of specific T-lymphocytes [25, 26]. However, further studies on a larger group with a similar course of COVID-19, including not only humoral, but also a cellular response to the infection, are necessary to reliably determine the immune response to SARS-CoV-2 infection in children with secondary immunodeficiencies.