DISCUSSION
CVID is characterized by recurrent and chronic respiratory tract infections and a broad spectrum of non-infectious complications, including chronic lung disease. Therefore, patients with CVID are at high risk of severe COVID-19 associated with poor outcomes. Despite an impaired specific antibody response to protein as well as polysaccharide antigens in patients with CVID , T-cell-mediated immunity is intact in most patients with CVID . Therefore, patients with CVID may benefit from vaccination against COVID-19, which may induce a specific T-cell response .
Owing to the COVID-19 outbreak, issues regarding immunogenicity, safety, and efficacy of vaccination in patients with IEI have been raised. However, studies on the immunogenicity and safety of vaccination in a broad spectrum of IEI have shown encouraging results. Amodio et al. revealed humoral and cellular responses in 86% and 76% of 21 patients with IEI, respectively, with no correlation with patient age , which is in contrast to the study by Hagin et al. Both authors also reported a lower humoral response than that in the general population, and only 4 (out of 12 patients) did not develop a cellular response . Similar findings were reported by Delmonet et al., in which specific anti-SARS antibodies were detected in 63 of 74 patients with IEI (85.1%). Furthermore, Leuween et al. demonstrated a negative correlation between the presence of non-infectious complications and immunosuppression in a large study of 505 patients, including 196 patients with CVID; however, these patients were vaccinated with the mRNA-1273 vaccine . Another significant limitation of previously published studies is the lack of prospective follow-up and limited vaccine-safety data in patients with IEI.
In the present study, a specific antibody response was observed in 52.4% of patients 1-month post-vaccination, and anti-RBD SARS-CoV-2 antibody levels were comparable to those in HCs. Nevertheless, seemingly favorable humoral responses differed significantly in qualitative properties and persistence over time. The neutralizing-antibody titer, which is predictive of the protection level , suggested a qualitative insufficiency, which is consistent with a previous study reporting a reduced capacity to produce virus-neutralizing antibodies in CVID . Additionally, the humoral response was not influenced by baseline or previous immunosuppression, or the presence of non-infectious complications. However, this might have been limited by the small number of patients included in our study. Responders were further characterized by a lower age (<40 years) and higher proportion of class-switched B cells, which is consistent with a previous study reporting an increased number of CD21low B cells, suggesting a possible dysregulation in the immune response to vaccination . Moreover, we demonstrated that a higher serum IgM concentration is a novel potential positive-response predictor. However, we did not observe differences in CD4 and CD8 subsets between responders and non-responders.
Additionally, although a specific T-cell response was detected at month 1 in less than half of the patients with CVID in our study, the proportion of responders was not significantly different than that observed in HCs (73%). Importantly, the proportion of both patients with CVID and HCs with persistently measurable Spike-specific T cell responses remained the same at month 6. However, we excluded four and five patients with CVID from T-cell response analysis at months 1 and 6, respectively, due to an absent anti-CD3 response and/or low viability which might have been related to T-cell abnormalities in CVID . Based on the immunophenotyping findings and central role of APRIL and BAFF in the survival and maturation of B cells and their dysregulation in CVID , we examined the serum concentration of both cytokines as potential response markers; however, we did not observe any significant differences in their levels between responders and non-responders at month 1. Additionally, no differences were found in serum IFNα concentrations, which can promote isotype switching .
Despite the promising antibody response rate at month 1, the proportion of responders rapidly decreased to 44.4% and 33.3% at 3 and 6 months, respectively. Anti-SARS-CoV-2 antibody titers were also significantly lower than those in HCs at the end of the study. The results were not influenced by IRT, and no specific anti-RBD antibodies were detected in IRT solutions. Noteworthily, previous studies used different methods to assess humoral and cellular responses. Therefore, our findings must be compared and interpreted with caution. Moreover, the specific anti-SARS antibody level that can predict immune protection remains unknown, and the efficacy of vaccination needs to be confirmed by long-term observation.
While a high vaccine-efficacy level has been observed in the general population, in our study, 3 of 21 patients (14.3%) tested positive for SARS-CoV-2 infection by RT-PCR. All patients were infected 6 months after second-dose administration when the SARS-CoV-2 Delta variant prevailed (November 21). The infected patients had a mild course and none developed a T-cell immune response, and the humoral response persisted in a single patient upon infection. Therefore, our data support booster vaccination in intervals shorter than 6 months for patients with CVID, as recommended for the general population .
Moreover, studies on vaccination safety in patients with IEI showed a favorable vaccination profile. In our study, AEs occurred in all patients with CVIDs, including injection-site reactions, fatigue, headaches, and fever. No SAE was reported. The spectra of the reactions after the first and second doses were comparable. Notably, we did not observe any changes in coagulation, including D-dimers, as a higher risk of thromboembolic events was described 15–21 days after BNT162b2 vaccine administration . Among the tested parameters, only soluble CD25 was significantly increased in patients with CVID post-vaccination.
In conclusion, to the best of our knowledge, this is the first study to investigate the long-term persistence of post-vaccination responses and clinical outcomes, providing data on a 6-month follow-up. We revealed that the anti-SARS-CoV-2 mRNA vaccine BNT162b2 induces a humoral response in a high proportion of patients with CVID. However, the vaccine induces lower anti-SARS-CoV-2 neutralizing antibody levels in patients with CVID than in the general population. Importantly, the antibody response was not persistent and continuously decreased 3 months after vaccination, whereas the CD4+ T-cell response persisted. We also demonstrated satisfactory clinical outcomes after vaccination in patients with CVID. No SARS-CoV-2 infections were reported within 5 months of the follow-up period, and only three patients (14.3%) tested positive for COVID-19; however, these patients had mild symptoms. Therefore, the BNT162b2 vaccine has a favorable safety profile in a proportion of patients with CVID, and this study supports booster vaccination in intervals shorter than 6 months for patients with CVID.