Efficacy of COVID-19 mRNA vaccination in patients with autoimmune disorders: humoral and cellular immune response

Background The impact of immunosuppressive therapies on the efficacy of vaccines to SARS-CoV-2 is not completely clarified. We analyzed humoral and T cell-mediated response after COVID-19 mRNA vaccine in immunosuppressed patients and patients with common variable immunodeficiency disease (CVID). Patients We enrolled 38 patients and 11 healthy sex- and age-matched controls (HC). Four patients were affected by CVID and 34 by chronic rheumatic diseases (RDs). All patients with RDs were treated by corticosteroid therapy and/or immunosuppressive treatment and/or biological drugs: 14 patients were treated with abatacept, 10 with rituximab, and 10 with tocilizumab. Methods Total antibody titer to SARS-CoV-2 spike protein was assessed by electrochemiluminescence immunoassay, CD4 and CD4-CD8 T cell-mediated immune response was analyzed by interferon-gamma (IFN-gamma) release assay, the production of IFN-gamma-inducible (CXCL9 and CXCL10) and innate-immunity chemokines (MCP-1, CXCL8, and CCL5) by cytometric bead array after stimulation with different spike peptides. The expression of CD40L, CD137, IL-2, IFN-gamma, and IL-17 on CD4 and CD8 T cells, evaluating their activation status, after SARS-CoV-2 spike peptides stimulation, was analyzed by intracellular flow cytometry staining. Cluster analysis identified cluster 1, namely the "high immunosuppression" cluster, and cluster 2, namely the "low immunosuppression" cluster. Results After the second dose of vaccine, only abatacept-treated patients, compared to HC, showed a reduced anti-spike antibody response (mean: 432 IU/ml +/- 562 vs mean: 1479 IU/ml +/- 1051: p = 0.0034), and an impaired T cell response, compared with HC. In particular, we found a significantly reduced release of IFN-gamma from CD4 and CD4-CD8 stimulated T cells, compared with HC (p = 0.0016 and p = 0.0078, respectively), reduced production of CXCL10 and CXCL9 from stimulated CD4 (p = 0.0048 and p = 0.001) and CD4-CD8 T cells (p = 0.0079 and p = 0.0006). Multivariable General Linear Model analysis confirmed a relationship between abatacept exposure and impaired production of CXCL9, CXCL10, and IFN-gamma from stimulated T cells. Cluster analysis confirms that cluster 1 (including abatacept and half of rituximab treated cases) showed a reduced IFN-gamma response, as well as reduced monocyte-derived chemokines All groups of patients demonstrated the ability to generate specific CD4 T activated cells after spike proteins stimulation. After the third dose of vaccine, abatacept-treated patients acquired the ability to produce a strong antibody response, showing an anti-S titer significantly higher compared to that obtained after the second dose (p = 0.0047), and comparable with the anti-S titer of the other groups. Conclusions Patients treated with abatacept showed an impaired humoral immune response to two doses of COVID-19 vaccine. The third vaccine dose has been demonstrated to be useful to induce a more robust antibody response to balance an impaired T cell-mediated one. All patients, exposed to different immunosuppressive drugs, were able to produce specific CD4-activated T cells, after spike proteins stimulation.