Coronavirus disease 2019 (COVID-19) is a respiratory infection caused by the β-coronavirus severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and the cause of the current pandemic (1). By December 2023, nearly 700 million cases have been registered, including more than 6.9 million deaths (2). Since the introduction of mass vaccination in December 2020, overall mortality and morbidity have been dramatically reduced (3). However, due to the emergence of new SARS-CoV-2 variants that bypass vaccine-induced immunological protection, the risk of severe COVID-19 disease requiring hospital admission in vaccinated patients remains a clinical and public health concern (4–6).

The decline in protective immunity, attributed to waning levels of antibodies and T-cell immunity post-vaccination, and the emergence of novel SARS-CoV-2 variants, constitute the two main factors contributing to the substantial increase in vaccine breakthrough SARS-CoV-2 infections worldwide (5–8). Even if vaccination has proven to be associated with a significantly reduced risk of severe outcomes and death in individuals of all ages (9–13), SARS-CoV-2 breakthrough infections remain a significant cause of hospitalization (14). Exploring the immunological characteristics of breakthrough SARS-CoV-2 infections is crucial for understanding the dynamics of vaccine efficacy over time, the potential immune responses that may impact disease severity, and, eventually, optimizing vaccination strategies to ensure the protection of individuals at the highest risk of severe outcomes.

We aimed to investigate and compare immune cell subpopulations and the induced immune responses comparing vaccinated and non-vaccinated patients hospitalized due to severe COVID-19. We hypothesized that vaccinated patients admitted with severe COVID-19 have a non-favorable immune profile, which could indicate immune impairment compared to non-vaccinated patients. We also aimed to identify overall differences in the immunologic signatures of SARS-CoV-2 breakthrough infections, aiming to provide a foundation for subsequent investigations.

This study compared immunologic signatures between vaccinated and non-vaccinated patients hospitalized with COVID-19. The main findings were that vaccinated patients (cases) admitted with with vaccine breakthrough COVID-19 infections had lower concentrations of CD19 B-cells and naïve CD4 T-cells and a higher proportion of TCRγδ1 T-cells. Furthermore, a higher unstimulated immune cell release of IL-10 at admission was observed.

We observed a lower B-cell concentration in vaccinated patients than in non-vaccinated patients admitted with COVID-19. It is well known that B-cell subsets, such as memory and plasma cells, are needed to protect from reinfection during repeated SARS-CoV-2 exposure (23). A reduced humoral response following vaccination could contribute to impaired COVID-19 immunity and hence admission to hospital despite COVID-19 vaccination. B-cell concentration has been strongly correlated with SARS-CoV-2 antibody production in immunocompromised patients (24). However, this association has not been observed in immunocompetent patients (25, 26). Our results, showing decreased B-cell concentrations in vaccinated patients, may indicate a suboptimal B-cell function in these patients, which would potentially contribute to hospital admission despite vaccination.

T-cells are crucial in antiviral immunity and are essential in immunological protection from SARS-CoV-2 infection (27). We observed T-cell cytopenia, particularly CD4 T-cells, and a lower concentration of naïve CD4 T-cells in vaccinated patients than in non-vaccinated patients. Several studies have reported that a successful response to vaccination is associated with a solid antigen-specific CD4 T-cell response (28–30). On the contrary, a low concentration of T-cells, including naïve CD4 T-cells, has been associated with poor vaccination responses. Some studies have also reported that a low concentration of naïve CD4 T-cells is associated with more severe COVID-19 (31–33). Our findings of a lower concentration of naïve CD4 T-cells in vaccinated patients may indicate a decreased T-cell repertoire and a smaller pool of T-cells to generate SARS-CoV-2 specific T-cells. This would ultimately increase the risk of impaired SARS-CoV-2 immunity. T- cells have been observed to be an important part of the immune response against SARS-CoV-2, where T-cell responses have a robust cross-recognition of SARS-CoV-2 variants despite varying antibody responses (34). This could suggest that vaccinated patients with severe COVID-19 due to breakthrough infection may have a suboptimal T-cell-mediated immune response, which should be further explored. CD4 T-cells are essential to achieve optimal B-cell function, and our finding of significantly reduced naïve CD4 T-cell counts may contribute to or may be related to reduced B-cell function. A study has observed a positive correlation between SARS-CoV-2 naïve CD4 T cells and viral clearance, where viral clearance increased with increasing SARS-CoV-2 specific naïve CD4 T cells (35). The authors of this study suggest, in concordance with the conclusion of our T-cell findings, that low counts of SARS-CoV-2 specific naïve CD4 T cells could implicate an insufficient B-cell activation leading to a decreased viral clearance (35). We suggest that an insufficient B-cell activation due to a suboptimal T-cell response could potentially affect the overall humoral response, leading to SARS-CoV-2 breakthrough infections. However, the measurement of SARS-CoV-2 specific T-cells combined with an assessment of the indirect effect on B-cell activation is required to confirm this.

Most studies have focused on classical TCRαβ T-cells when describing the role of T-cells in COVID-19. In contrast, few studies have explored the role of TCRγδ T-cells in COVID-19 (36, 37). TCRγδ1 and TCRγδ2 are the two main TCRγδ subsets and the most studied. TCRγδ T-cells, a component with innate-like immune system properties, respond to inflammation and stressed or infected cells (38). In vitro, TCRγδ T-cells have also shown the ability to eliminate SARS-CoV-2 (39). TCRγδ T-cells are therefore considered an essential part of the innate defense against viruses, including SARS-CoV-2. TCRγδ2 T-cells can interact with other immune cells, including B-cells and dendritic cells, and carry out cytolysis through the effects of cytotoxic perforin and granzymes (40, 41).

Furthermore, a recent study observed an association between TCRγδ T-cells and disease severity, where a low frequency of TCRγδ2 T-cells, which usually are the dominant TCRγδ T-cells in adults, were associated with severe COVID-19 disease (42–44). In our study, we observed a markedly higher proportion of TCRγδ1 T-cells and a lower proportion of TCRγδ2 T-cells in vaccinated patients compared to non-vaccinated patients. Overall, these observations could indicate a suboptimal TCRγδ T-cell response in vaccinated patients, where previous studies have shown that TCRγδ T-cells are highly involved and play an essential role in the host immune response against SARS-CoV-2 (38). Our findings could indicate that the difference in TCRγδ T-cell proportions in vaccinated patients could contribute to a non-favorable immune cell profile, potentially leading to subsequent hospital admission. These findings may further underline the importance of T-cells during SARS-CoV-2 infection.

We assessed the ex vivo-induced immune response by TruCulture as a proxy for immune function. We observed a significant increase in the unstimulated release of IL-10 in vaccinated patients. This observation may reflect both high unstimulated ex vivo release of IL-10 during incubation and high circulating plasma levels of IL-10. IL-10 is a classical anti-inflammatory cytokine with immunosuppressive activities, downregulating various cytokines and co-stimulatory molecules (45). Several studies have reported that early production of IL-10 is associated with poor clinical outcomes in COVID-19 (46–48). A recent meta-analysis further concluded that circulating IL-10 can be used as a predictor for patients’ clinical status and survival, where IL-10 is suggested to be the main cause of the immunodepression associated with SARS-CoV-2 infection (49). However, further studies with larger study populations are needed to confirm our finding of higher ex vivo release of IL-10/higher circulating levels of IL-10 in patients with SARS-CoV-2 vaccine breakthrough infection and to determine the potential influence of this phenomenon on the COVID-19 disease course.

Research on IL-10 in the context of COVID-19 has thoroughly evaluated its role, revealing that higher levels of IL-10 correlate with more severe disease and have been shown to predict the progression to severe or critical disease (47, 48). These studies were larger and focused on the clinical impact of IL-10, rather than the immunological phenotype of severe COVID-19 stratified by vaccination status. In this study, we did not observe a difference in disease severity between vaccinated and non-vaccinated patients. The study’s aim was exploratory regarding immunological phenotype, and our findings should, therefore, be interpreted as indicative rather than conclusive. Further studies with a larger sample size assessing SARS-CoV-2 specific immune cells, antibodies, the effect of clinical variables on observed immune response differences and samples taken before COVID-19 (or a follow-up sample) are highly warranted to fully evaluate the underlying pathophysiology and potential causes regarding our observed immune response differences.

Our study has strengths and limitations. First, we present detailed, in-depth immunological profiles of patients hospitalized with severe COVID-19 at admission, providing valuable insights into the immune responses of vaccinated and non-vaccinated individuals otherwise underreported in the literature (50, 51). In contrast to other reports (48, 49), we did not only focus on specific immune cells but had a rather broad approach with data on cytokine profiles, B cells, T cells, and T cell receptor profiles, activation, and exhaustion. This very deep and broad immunophenotyping, paired with data on inflammation through cytokine assessment, is unique and scarce in the literature. We could not find similar studies regarding deep and broad immunophenotyping combined with cytokine responses stratified by vaccination status in patients hospitalized with severe COVID-19. We applied strict inclusion criteria, ensuring a well-defined study population and minimizing the potential risk of confounding. The observational nested case-control design facilitates comparing immunological characteristics between different patient groups, contributing to a better understanding of vaccine breakthrough infections.

We acknowledge that the results of our study are limited by a relatively small sample size, emphasizing the need for more extensive studies to validate findings. We consider this study a descriptive, hypothesis-generating study, with its strength in deep immunophenotyping and broad immunological analysis, coupled with strict sex and age case matching. Also, only patients with community-acquired SARS-CoV-2 infections were included. Thus, we aimed to optimize the comparison between vaccinated vs. non-vaccinated groups, which constituted the premise of our study. Despite our efforts, recruiting sex- and age-matched non-vaccinated patients hospitalized with COVID-19 was challenging, likely because of Denmark’s widespread acceptance of COVID-19 vaccination (52). This contributed, at least in part, to the limitations in the sample size. We also included two immunocompromised individuals in the cohort. Even though the number of individuals is minimal and no conclusions can be drawn, the detailed description is of value since this group of patients is likely to constitute a large majority of patients with breakthrough and persistent SARS-CoV-2 infections.

Finally, adding a third comparison group with vaccinated, non-hospitalized patients and antibody titers would have been ideal. Still, previous studies, including by our group, assessed neutralizing antibody titers for up to 6 months following primary COVID-19 infection. Although many patients may still have high neutralizing antibody titers (18), differences in antibody persistency and response to vaccination may contribute to the occurrence of breakthrough infections. The absence of SARS-CoV-2-specific immune cell stimulations and antibody measurements also limits our study. Furthermore, samples were taken during severe disease, challenging the interpretation of the results. In the future, studies should aim to overcome these challenges by, for instance, allowing the recruitment across multiple sites, including different groups of immunocompromised patients, thereby aiming for generalizability and robustness of findings.

Overall, the differences observed in this study between vaccinated and non-vaccinated patients admitted to hospitals with COVID-19 suggest that vaccinated patients display a non-favorable immune response, which previous studies have identified as potentially associated with immune impairment. We observed low total B-cell counts, low CD4 naïve T-cells, a skewed TCRγδ V1/V2 ratio, and an exaggerated IL-10 response in vaccinated compared to non-vaccinated patients. These observations align with previous studies, where similar findings in B-cells, T-cells, and release of/circulating IL-10 have been associated with a weak response to vaccination and more severe clinical disease (23, 31–33, 42, 48, 49, 53). We suggest a potentially suboptimal response to vaccination, perhaps due to an underlying degree of immune impairment, could contribute to the observed differences. However, further studies, including simultaneous serological analyses, are required to assess this. Continued research and further follow-up studies evaluating these immunological differences are warranted.

The datasets presented in this article are not readily available because data can be made available by the investigators after review of a suitable protocol by the steering committee of the study group upon request. Requests to access the datasets should be directed to Sisse.Rye.Ostrowski@regionh.dk; zitta.barrella.harboe@regionh.dk; Hanne.Marquart@regionh.dk.

The study was approved by the Danish Ethics Committee (H-20026502) and the Danish Data Protection Agency (P-2020-426). The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

AS: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Writing – original draft, Writing – review & editing. HH: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Project administration, Supervision, Writing – review & editing. JH: Supervision, Writing – review & editing. LK: Investigation, Project administration, Writing – review & editing. BL: Funding acquisition, Investigation, Resources, Supervision, Writing – review & editing. AD: Investigation, Supervision, Writing – review & editing. FG: Investigation, Project administration, Writing – review & editing. MM: Investigation, Methodology, Project administration, Writing – review & editing. RT: Investigation, Methodology, Project administration, Writing – review & editing. CN: Conceptualization, Funding acquisition, Investigation, Project administration, Resources, Supervision, Writing – review & editing. PB: Investigation, Project administration, Writing – review & editing. CJ: Conceptualization, Supervision, Writing – review & editing. KF: Conceptualization, Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing. TF: Conceptualization, Resources, Supervision, Writing – review & editing. HM: Conceptualization, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Writing – review & editing. ZH: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – original draft, Writing – review & editing. SO: Conceptualization, Formal Analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Writing – review & editing.