ABSTRACT
Background: Repeated vaccination against SARS-CoV-2 increases serological response in kidney transplant recipients (KTR) with high interindividual variability. Still, no decision support tool exists to predict SARS-CoV-2 vaccination response in KTR. Method(s): We developed, internally and externally validated five different multivariable prediction models of serological response after the third and fourth vaccine dose against SARS-CoV-2 in KTR. Using 27 candidate predictor variables, we applied statistical and machine learning approaches including logistic regression (LR), LASSO LR, random forest, and gradient boosted regression trees. For development and internal validation, data from 585 vaccinations were used. External validation was performed in four independent, international validation datasets comprising 191, 184, 254, and 321 vaccinations, respectively. Result(s): Internal validation using a rigorous resampling approach showed AUC-ROC of 0.825 for LASSO LR, which was then used for model fitting and external validation. LASSO LR performed on the whole development dataset yielded a 23- and 11-variable model, respectively. External validation showed ROC-AUC of 0.855, 0.749, 0.828, and 0.763 for the sparser 11-variable model, yielding an overall AUC-ROC of 0.809, and a negative predictive value of 0.752. The 23-variable model showed AUC-ROC of 0.853, 0.714, 0.844, and 0.778 in four independent validation sets, yielding an overall AUC-ROC of 0.818, and a negative predictive value of 0.795. Conclusion(s): Both, an 11- and 23-variable LASSO LR model predict vaccination response in KTR with good AUC-ROC. Implemented as an online tool at https://www.txvaccine. com, it can guide decisions when choosing between different immunization strategies to improve protection against COVID-19 in KTR. (Figure Presented).
ABSTRACT
Purpose: Mortality from COVID-19 among kidney transplant recipients (KTR) is unacceptably high, and their response to up to three vaccinations against SARSCoV- 2 is strongly impaired. We provide the first systematic analysis of serological response to up to five repeated vaccinations in nonresponding KTR. Method(s): We retrospectively analyzed serological response to basic immunization, as well as administration of three, four and five doses of SARS-CoV-2 vaccine in KTR from December 27, 2020 until December 31, 2021. In particular, the influence of different dose adjustment regimens for mycophenolic acid (MPA) on serological response to fourth vaccination was analyzed. Result(s): In total, 4.277 vaccinations against SARS-CoV-2 in 1.478 patients were analyzed. Serological response was 19.5% after 1.203 basic immunizations, and increased to 29.4%, 55.6%, and 57.5% after 603 third, 250 fourth and 40 fifth vaccinations, resulting in a cumulative response rate of 88.7% (figure 1A-B). Patients with belatacept immunosuppression show impaired cumulative serological response (4.4%, 12.4%, and 16.4%) in comparison to patients with calcineurin inhibitor (CNI)- based immunosuppression (19.1%, 37.6%, and 70.1%) after basic immunization, three, and four vaccinations (figure 1C-F). In patients with CNI and MPA maintenance immunosuppression, pausing MPA and adding 5 mg prednisolone equivalent before the fourth vaccination increased serological response rate to 75% in comparison to no dose adjustment (52%) or dose reduction (46%) without occurence of rejections (figure 2). Conclusion(s): Repeated SARS-CoV-2 vaccination of up to five times effectively induces serological response in kidney transplant recipients. It can be enhanced by pausing MPA at the time of vaccination. Patients with belatacept immunosuppression are unlikely to achieve sufficient serological response and require different approaches.
ABSTRACT
Systemic vaccination against SARS-CoV-2 elicited high titers of specific antibodies in the blood and in the oral cavity. Preexisting autoimmune diseases, such as rheumatoid arthritis, and biological treatments, like B cell depletion, are known to exhibit higher risk of severe COVID-19 manifestation and increased frequency of breakthrough infections after vaccination. We hypothesized that such increased risk is associated with an aberrant induction of secreted antibodies in the oral cavity. Here we evaluated the levels of secreted antibodies in the oral cavity against the SARS-CoV-2 Spike protein during the course of vaccination in RA patients with or without B cell depletion. We found that total salivary IgG levels were correlated with number of B cells in the blood. Anti-Spike IgG responses 7 days after second vaccination were induced in the oral cavity of all healthy individuals, while only 6 out 23 RA patients exhibited anti-Spike IgG in their saliva regardless of B cell depleting therapy. Importantly, both salivary and serologic anti-Spike IgG and IgA responses towards WT and omicron Spike variants were efficiently induced by third vaccination in RA patients with or without B cell depletion to the levels that were similar to healthy individuals. Altogether, these data advocate for the necessity of three dose vaccination for RA patients to mount anti-Spike antibody responses at the mucosal surfaces and annotate the reduction of secreted salivary IgG by B cell depletion.
ABSTRACT
Background: Vaccination is considered efficient in controlling infections incl. SARS-CoV-2. Prior studies showed that patients receiving rituximab (RTX) with low B cell counts are at increased infectious risk (1) and risk of inadequate vaccination responses (2, 3). Thus, the ability to further defne and predict vaccination responses in these patients may guide their optimal protection. Objectives: To assess predictive biomarkers of vaccination responses upon SARS-CoV-2 vaccination in RTX treated patients. Methods: B cell characteristics before vaccination were evaluated to predict responses in 15 patients with autoimmune infammatory rheumatic diseases receiving RTX. 11 patients with rheumatoid arthritis on other therapies (RA), 11 kidney transplant recipients (KTR) and 15 healthy volunteers (HC) served as controls. A multidimensional analysis of B cell subsets and a correlation matrix were performed to identify predictive biomarkers. Results: Signifcant differences regarding absolute B cell counts and specifc subset distribution pattern between the groups were validated at baseline. Here, the majority of B cells from vaccination responders of the RTX group (RTX IgG+) comprised naïve and transitional B cells, whereas vaccination non-responders (RTX IgG-) carried preferentially plasmablasts and double negative (CD27-IgD-) B cells (Figure 1). Moreover, there was a positive correlation between neutralizing antibodies and absolute B cell numbers with B cells expressing HLA-DR and CXCR5 (involved in antigen presentation and germinal center formation) as well as an inverse correlation with CD95 expression and CD21low expression (marker for activation and exhaustion) on B cells. Conclusion: Substantial repopulation of naïve B cells upon RTX therapy appears to be essential for an adequate vaccination response requiring germinal center formation. In contrast, expression of exhaustion markers (CD21low, CXCR5-, CD95+) indicate negative predictors of vaccination responses. These results may guide optimized vaccination strategies in RTX treated patients clearly requiring antigen-inexperienced B cells for appropriate protection.