Diminished seroconversion following a single SARS-COV-2 vaccine in ocrelizumab-treated relapsing-remitting multiple sclerosis patients

Background: Despite impressive efficacy in immunocompetent individuals, the immunogenicity of a single dose of COVID-19 vaccine in B-cell-deplete patients remains unknown. Objectives: We aimed to quantify real-world vaccine immunogenicity in ocrelizumab recipients. Methods: We measured post-vaccination SARS-COV-2 immunoglobulin G (IgG) in ocrelizumab recipients using a highly sensitive Luminex assay. Results: 44.1% of patients had detectable SARS-COV-2-IgG 21+ days after one vaccine dose, regardless of vaccine type (AZD1222 vs BNT162b2, odds ratio (OR) = 0.62, 95% confidence interval (CI) = 0.157–2.32, p = 0.72). B-cell count strongly predicted seroconversion (β1 = 12.38, 95% CI = 4.59–20.16, p = 0.0029), but undetectable B-cells did not preclude it. The second vaccine seroconverted 53% of the patients who had not already responded to dose 1. Conclusion: Humoral response after one COVID-19 vaccine dose is lower than expected in CD20-deplete patients.


Introduction
Patients taking immunosuppressive medication have been prioritised for SARS-COV-2 vaccination. 1 The efficacy of several SARS-COV-2 vaccines has been demonstrated in immunocompetent individuals, emphasising humoral immunity outcomes. 2,3 However, the efficacy of a single vaccine dose in B-cell-depleted patients remains unknown. To establish whether they mount a humoral immune response, we measured SARS-COV-2 IgG seroconversion post-vaccination.

Methods
We audited the outcomes of SARS-COV-2 vaccination in individuals treated with ocrelizumab, as per section 4.4 of the ocrelizumab SmPC 4 and UK government 'Green Book' guidance. 5 Between January and April 2021, all patients receiving ocrelizumab for relapsing-remitting MS at a single centre were invited for blood tests as part of medical care, including extended lymphocyte phenotyping, immunoglobulins and SARS-COV2-IgG. Patients received either the BNT162b2 or AZD1222 SARS-COV2 vaccine via the UK National Health Service as standard medical care for people with MS. Patients were asked to give blood at baseline (within ±3 days of vaccine dose 1) and again 21+ days after the first vaccine. Individuals who were seronegative after one vaccine dose were offered a repeat SARS-COV-2 IgG test after the second dose.
The Luminex-based SARS-COV2-IgG assay has been previously described. 6 Briefly, carboxylated beads covalently coupled to SARS-COV-2 peptides capture human IgG against SARS-COV-2 receptor-binding domain (R), trimeric spike protein (S) and nucleocapsid protein (N). Beads were incubated with patient sera (1/100 dilution) for 1 hour in 96-well filter plates (MultiScreenHTS; Millipore) at room temperature in the dark on a horizontal shaker. Beads were washed three times with 10 mM PBS/0.05% Tween 20, incubated for 30 minutes with a Phycoerythrin (PE)labelled anti-human IgG-Fc antibody (Leinco/ Biotrend), washed again and re-suspended in 100 μL PBS/Tween. They were analysed on a Luminex analyser (Luminex/R&D Systems) using Exponent Software V31. Positivity was defined by receiver operating characteristic (ROC) analysis as S-antibody levels exceeding 1896 median fluorescence intensity (MFI); the threshold for R-antibody was 456 MFI, and for N-antibody was 6104 MFI. This assay has been validated on patients with polymerase chain reaction (PCR)-confirmed SARS-COV-2 infection, in whom it is 84% sensitive and 100% specific. Cross-reactivity with healthy control pre-pandemic sera is very rare (due to similar N-protein in other Coronaviridae). Immunocompetent vaccinated patients typically test positive for anti-S and/or receptor-binding domain (RBD), but not anti-N antibodies. Previous infection results in dual-positive anti-S and N antibodies.
Lymphocyte counts were measured by standard clinical flow cytometry on whole blood, contemporaneous with pre-vaccine SARS-COV-2 IgG testing. Data were analyzed in the GraphPad Prism v8.3.0 using the two-tailed Mann-Whitney or Fisher's exact test. Results were considered significant at alpha = 0.05, equivalent to a Bonferroni-adjusted p < 0.0125; we show unadjusted p-values. Exploratory analysis was performed using the least-squares multiple logistic regression. Candidate models were assessed for goodness-of-fit using the Akaike Information Criterion (AICc). Data are presented as mean value ± standard error of the mean (SEM) unless specified.

Results
Patient characteristics and numbers are summarised in Table 1.

Discussion
We aimed to determine if patients treated with the B-cell-depleting drug ocrelizumab produce a detectable humoral immune response following SARS-COV-2 vaccination. We observed an attenuated early humoral immune response in this population following one vaccine dose.
44.1% of ocrelizumab-treated patients developed SARS-COV2-IgG 21+ days after dose 1 of the vaccine. By contrast, most phase 1/2 trial participants receiving one dose of BNT162b2 vaccine had detectable anti-S IgG at day 21 (11/11 aged = 18-55 years; 11/12 aged = 65-85 years). 7 91%-100% of participants developed neutralising antibodies 28 days after one dose of AZD1222 8 and 79% of patients developed neutralising antibodies after one BNT162b dose. 9 Here, we measured the presence of S-, N-and/or RBD-antibodies (not neutralisation). Even accounting for false-negatives (16%), we observed a diminished rate of SARS-COV-2-IgG seroconversion after 1 vaccine dose. However, the second vaccine generated antibodies in 53% of the patients who had originally not developed detectable antibodies after one dose.
Our findings are consistent with previous observations of diminished immunity to tetanus toxoid and pneumococcal antigen in ocrelizumab recipients. 10 A controlled study in 23 ocrelizumab recipients showed humoral immunity in only 22% patients after two SARS-COV-2 vaccine doses. 11 In our cohort, detectable B-cells strongly predicted SARS-COV-2 seroconversion, but the absence of circulating B-cells did not preclude it. Furthermore, T-cell-mediated immunity is likely to be an important unmeasured protective factor. Vaccination with AZD1222 produces a peak Th-1 skewed response 14 days after one dose, 8 while RBD-specific CD4+ Th1-cells are detectable 7 days after the second dose of BNT162b1. 12 This should not differ substantially in B-cell deplete individuals.
This study adds to mounting evidence for diminished early humoral immunity in CD20-deplete patients. The contribution of cellular immunity and booster vaccination remains to be fully assessed.