key: cord-0857610-0ervxt84
authors: Madelon, Natacha; Lauper, Kim; Breville, Gautier; Sabater Royo, Irène; Goldstein, Rachel; Andrey, Diego O; Grifoni, Alba; Sette, Alessandro; Kaiser, Laurent; Siegrist, Claire-Anne; Finckh, Axel; Lalive, Patrice H; Didierlaurent, Arnaud M; Eberhardt, Christiane S
title: Robust T cell responses in anti-CD20 treated patients following COVID-19 vaccination: a prospective cohort study
date: 2021-11-17
journal: Clin Infect Dis
DOI: 10.1093/cid/ciab954
sha: 07d833a2e6e25f7a193ee3df64f0f5340b613ec7
doc_id: 857610
cord_uid: 0ervxt84

BACKGROUND: Patients treated with anti-CD20 therapy are particularly at risk of developing severe COVID-19, however little is known regarding COVID-19 vaccine effectiveness in this population. METHODS: This prospective observational cohort study assesses humoral and T-cell responses after vaccination with 2 doses of mRNA-based COVID-19 vaccines in patients treated with rituximab for rheumatic diseases or ocrelizumab for multiple sclerosis (n=37), compared to immunocompetent individuals (n=22). RESULTS: SARS-CoV-2-specific antibodies were detectable in only 69.4% of patients and at levels that were significantly lower compared to controls who all seroconverted. In contrast to antibodies, Spike (S)-specific CD4+ T cells were equally detected in immunocompetent and anti-CD20 treated patients (85-90%) and mostly of a Th1 phenotype. Response rates of S-specific CD8 (+) T cells were higher in ocrelizumab (96.2%) and rituximab-treated patients (81.8%) as compared to controls (66.7%). S-specific CD4 (+) and CD8 (+) T cells were polyfunctional but expressed more activation markers in patients than in controls. During follow-up, three MS patients without SARS-CoV-2-specific antibody response had a mild breakthrough infection. One of them had no detectable S-specific T cells after vaccination. CONCLUSIONS: Our study suggests that patients on anti-CD20 treatment are able to mount potent T-cell responses to mRNA COVID-19 vaccines, despite impaired humoral responses. This could play an important role in the reduction of complications of severe COVID-19.

In patients with immune-mediated rheumatic diseases (RD) and multiple sclerosis (MS), immunosuppressive drugs and in particular anti-CD20 therapy are associated with an increased risk of severe COVID-19 [1] [2] [3] . Although generally identified as priority groups for vaccination, these patients were not included in pivotal studies evaluating the efficacy of COVID-19 vaccines and their effectiveness in this population is still unknown. Anti-CD20 treatment depletes B cells and impairs antibody responses to classical vaccines [4] [5] [6] . Several studies have now confirmed reduced antibody levels and seroconversion rates in anti-CD20 treated patients following SARS-CoV-2 infection [7] and COVID-19 vaccination, irrespective of the underlying disease [8] [9] [10] [11] . Although antibodies are likely to play a critical role in preventing infection, recovery from COVID-19 in patients with X-linked agammaglobulinemia suggests that antibodies are not mandatory to overcome disease [12] . T cells may also be involved in protection against COVID-19 [12] [13] [14] and memory T cells are readily detectable several months after infection [15] . As B cell could play a role as antigenpresenting cells to naïve T cells, the question remains as whether B-cell depleted patients could still mount functional T cell responses to COVID-19 vaccines, which may provide some level of protection against severe disease.

The aim of our study was thus to characterise and compare T cell responses to mRNA-based COVID-19 vaccines between patients with rheumatic diseases and multiple sclerosis treated with anti-CD20 therapy and immunocompetent controls.

We included individuals ≥ 18 years of age scheduled to receive the COVID-19 vaccine or having received ≤ 2 COVID-19 vaccine doses in the last 5 weeks. Subjects with SARS-CoV-2 documented infection less than 3 months prior to inclusion or ongoing signs of febrile or non-febrile infection were excluded. Further details in the study are in supplementary methods.

A c c e p t e d M a n u s c r i p t 5 

This prospective observational study was conducted at the Geneva University Hospitals (HUG), Switzerland according to the principles of Good Clinical Practice and was approved by the Geneva Cantonal Ethics Commission (2021-00430). Informed consent was obtained from all participants.

Antibodies were measured in sera using the Elecsys Anti-SARS-CoV-2 nucleoprotein (anti-N total antibodies) and Anti-SARS-CoV-2 Spike (anti-receptor binding domain (RBD) total antibodies) on the Cobas e801 analyzer (Roche Diagnostics, Switzerland). Seroconversion was defined as > 0.8 IU/ml for the anti-RBD antibodies and > 1 cut-off index (COI) for anti-N antibodies as previously defined [16] . For the AIM assay, PBMC were stimulated with 1ug/ml of SARS-CoV-2 megapool peptides (CD4-S, 15-mer peptides overlapping by 10 amino acids spanning the entire antigen, n=253) or in DMSO (negative controls). For intracellular cytokine production, Brefeldin A (Golgiplug, BD) was added to the culture overnight. Cells were stained with specific antibodies and analysed by flow cytometry (see supplementary methods for further details). Percentage of S-specific T cells, AIM+, cytokines+ or granzyme B + cells were calculated by subtracting the value of the corresponding DMSO stimulation control samples.

Statistical analysis was performed in GraphPad Prism software (version 8.0.2). Kruskal-Wallis test with Dunn's multiple comparisons was used (unless otherwise stated in the figure legends) and categorical variables were compared using Fisher's exact test (3x2). Correlation analyses were performed using Spearman test. P values from correlations were corrected for multiple comparisons using the False Discovery Rate method, and p < 0.05 was considered statistically significant.

In order to assess the effect of B cell depletion on vaccine-induced T cell responses, we studied a total of 37 patients treated with either ocrelizumab (n=26) for multiple sclerosis (MS) or rituximab (n=11) for rheumatic diseases (RD) compared to 22 age-matched immunocompetent controls (details see Table 1 ). Most patients with RD were treated for rheumatoid arthritis (n=7) and had more co-morbidities compared to MS patients or controls.

A c c e p t e d M a n u s c r i p t 6 Mean age was balanced between groups. Regarding concomitant medication, 5/11 RD patients received another immunosuppressor (mainly methotrexate or corticosteroids), while all MS patients were treated with ocrelizumab only. Hence, this cohort is likely to reflect the effect of short-to long-term treatment with anti-CD20 on vaccine response in the absence of other major confounding factors, in particular age and concomitant immunosuppressive drugs. The interval between last anti-CD20 treatment and 1 st vaccine dose was shorter in patients under ocrelizumab (median 24.9 weeks) than in patients treated with rituximab (median 42 weeks), which explained the absence or low (<2 %) percentage of CD19+ B cells among total PBMC at time of vaccination, especially in ocrelizumab-treated patients (Sup Fig   1) . Most patients (34/37, 91.9%) and a majority of controls (15/22, 68 .2%) had no history of SARS-COV-2 RT-PCR-confirmed infection (review of medical record) prior to vaccination (Table 1 ) nor a positive anti-nucleoprotein serology (measured 30 days after vaccination, Fig   1A) . Fig 1B) , which were lower in patients on ocrelizumab due to a more recent treatment. There was no correlation between age and antibody response. The three patients with a known history of COVID-19 or with detectable anti-N antibodies did not have higher antibody responses as compared to those unexposed ( Fig 1A, Sup Fig 2A) . 

T-cell immunity against SARS-CoV-2 is thought to play a role in protection against severe disease [14] . To assess if mRNA vaccines could elicit T cell responses in our patient cohort as reported for healthy individuals [18, 19] , we stimulated PBMC collected 30 days after the second vaccine dose with a pool of peptides covering the S-protein [17] and identified Sspecific T cells using the activation-induced marker (AIM) assay. S-specific OX40+ 41-BB+ CD4+ T cells were equally induced in immunocompetent and anti-CD20 treated patients ( Fig   1C) , with a high frequency of responders (85-91%). S-specific CD69+ 41BB+ CD8+ T cells were detectable at similar levels in all groups (Fig 1E) , however there was a statistically significant higher response rate found in ocrelizumab-(96.2%; 25/26) and rituximab-treated patients (81.8%; 9/11) compared to controls (66.7%; 14/21, p=0.02). Previous history of COVID-19, the type of mRNA vaccine and time since last anti-CD20 treatment had no impact on the level of S-specific T-cells and results were similar when participants with an history of COVID-19 were excluded from the analysis (data not shown). Interestingly, there was a weak inverse correlation between the magnitude of S-specific CD8+ T cells and anti-RBD antibody responses considering all participants (R= -0.32, p=0.049, Fig 1F , not significant for CD4+ T cells, Fig 1D) . Lastly, the higher frequency of patients with AIM+ CD8+ T cells compared to controls was probably not due to higher levels of pre-existing cross-reactive T cells: in a subset of previously uninfected patients (n=13), S-specific AIM+ CD4+ and CD8+ T cells were undetectable at time of first vaccination (Sup Fig 2B, C) .

We then assessed the functionality of antigen-specific T cells to understand if the quality of T cell responses is altered in the absence of B cells. mRNA COVID-19 vaccines are known to predominantly induced Th1 CD4+ T cells expressing IL-2, IFN-gamma and the transcription factor Tbet rather than Th2 (IL-13+, GATA3+) or Th17 (IL-17+, RORgammaT+) cells [20] .

First, we observed that a majority of S-specific AIM+ CD4+ T cells expressed Tbet in all patients and controls and found that some ocrelizumab-treated patients had a higher percentage of GATA-3+ T cells, however not reaching statistical significance at the group level (Fig 2A) . Next, we used intracellular cytokine staining to evaluate if S-specific T cells express several cytokines, given that polyfunctional T cells are often associated with improved vaccine-induced protection to viral infection. We found that anti-CD20 treated patients had similar level of S-specific CD4+ T cells expressing at least 2 of the markers IL-A c c e p t e d M a n u s c r i p t 8 2, TNF-alpha, IFN-gamma or granzyme B as compared to controls, suggesting a similar polyfunctionality ( Fig 2B) . The frequency of S-specific CD4+ T cells producing IL-2 and IL-2+TNF-alpha+ in both ocrelizumab and rituximab-treated patients was however higher as compared to immunocompetent controls, while the percentage of CD4+ T cells expressing IFN-gamma alone or in combination with other cytokines was similar (Fig 2C and Sup Fig   3A) . There were no detectable IL-13 or IL-17-expressing CD4+ T cells (Sup Fig 3A) .

We next assessed whether S-specific CD8+ T cells were also functional in our patient cohorts. S-specific CD8+ T cells expressing either IL-2 or IFN-gamma were detected in more patients treated with anti-CD20 than in controls (Fig 3A) . The percentage of IL-2-expressing cells was significantly higher (p=0.013) in ocrelizumab-treated patient as compared to controls while only a trend was observed for IFN-gamma (p=0.07) and for rituximab-treated patients for both cytokines. Similar to CD4+ T cells, patients on anti-CD20 treatment had polyfunctional S-specific CD8+ T cells, with a trend for more cells expressing at least 3 markers and significantly more single IL2+ cells than controls (Fig 3B, C) . In general, a higher frequency of S-specific CD8+ T cells co-expressing granzyme and cytokines were found in patients as compared to controls.

Finally, we assessed the memory phenotype of S-specific AIM+ CD4+ and CD8+ T cells and did not find any difference between groups ( Sup Fig 3 and 4 ). CD8+ T cells had predominantly an effector memory phenotype (CD45RA-CCR7-) while CD4+ T cell phenotypes were equally effector (CD45RA-CCR7-) and central memory (CD45RA-CCR7+).

Altogether, our data suggest that S-specific T cells induced by mRNA vaccines have a similar functional and memory profile but a more activated phenotype in anti-CD20 treated patients as compared to controls. Patients under anti-CD20 therapy had lower levels of RBD-specific antibodies as compared to controls, in line with what was reported by others and as expected from experience with other vaccines [10, 11, 21] . In general, more patients treated with Rituximab had higher CD19+ B cells counts than patients treated with ocrelizumab, likely due to a longer interval between last treatment and vaccination in the RD group. However, some patients developed RBD-specific antibodies and these responses correlated with levels of circulating CD19+ B cells measured at day 30 after vaccination, but not with time since last treatment (data not shown). It is unclear at this stage whether the level and neutralizing capacities of the antibodies in those patients who seroconverted will be sufficient to prevent infection or severe COVID-19.

We found that patients under anti-CD20 therapy with a known history of COVID-19 or with detectable anti-N antibodies did not have higher antibody responses as compared to those unexposed. This suggests that on anti-CD20 treatment, previous exposure to SARS-CoV-2 does not provide an advantage in terms of humoral vaccine response, in contrast to what we ( Fig 1A) and others [22] observed in immunocompetent individuals.

Although it is not yet demonstrated that T-cell immunity against SARS-CoV-2 play a direct role in protection against severe disease in humans (but in animals [14] ), it may provide some level of protection to vaccinated patients under anti-CD20 treatment despite their limited antibody response. The number and functionality of T cells is generally maintained after treatment with B-cell targeting drugs, although depletion of some CD20+ T cells, an increase in memory and loss of terminally differentiated CD4+ T cells have been reported [23, 24] .

Despite lower antibody responses, anti-CD20 treated patients mounted robust S-specific CD4+ and CD8+ T cell responses following COVID-19 mRNA vaccination, a finding similar to a recent study measuring pan-T-cell responses by IFNgamma-ELISpot in rituximab-treated patients [25] . In line with our results, SARS-CoV-2-specific CD4+ and CD8+ T cells are also detectable in exposed family members and in convalescent patients with asymptomatic/mild COVID-19 who remained seronegative [26] .

Surprisingly, patients on anti-CD20 therapy developed strong S-specific CD8+ T cell responses and presented higher response rates compared to controls. The level of S-specific CD8+ T cells was inversely associated with anti-RBD antibody responses. One hypothesis to explain higher T cell activation in those patients could be the presence of more activated APCs (eg. monocytes) at time of vaccination as a result of B cell depletion [27] . Interestingly, in hematologic cancer patients treated with anti-CD20, a greater number of CD8+ T cells is associated with improved survival to COVID-19, despite impairment in humoral immunity, and 77% of patients had detectable SARS-CoV-2-specific T cell responses [28] .

Additionally, it has been recently shown that robust and functional CD8+ T cell responses are elicited only one week after the BNT162b2 prime vaccination when neutralizing antibodies are weakly detected [29] and at a time when protective effect of COVID-19 mRNA vaccines can be observed [30, 31] . The question remains whether the strong CD8+ T cell responses elicited in anti-CD20 treated patients would be sufficient to prevent severe COVID-19. All Patients on anti-CD20 had polyfunctional S-specific CD4+ and CD8+ T cells, with more T cells producing at least 2 or 3 cytokines. In addition, the frequency of S-specific CD4+ and CD8+ T cells producing IL-2 in patients under anti-CD20 therapy was higher as compared to controls. Finally, rituximab and ocrelizumab-treated patients developed antigen-specific T cell memory responses similar to immunocompetent controls. CD8+ T cells had predominantly an effector memory phenotype while CD4+ T cell phenotypes were equally effector and central memory, as described by others following infection [15] . This suggest that in addition to generating polyfunctional T cells, it is likely that mRNA vaccines can generate long-lasting memory T cells in patients treated with anti-CD20 as shown for healthy individuals [32] . This will be confirmed in a follow-up of the current study by looking at the persistence of the cellular response in these patients at 6 and 12 months post vaccination.

Limitations of our study include the small sample size and the short follow-up after vaccination. We also were not able to correlate T cell findings with clinical protection as the study was not designed to measure efficacy, which is pivotal in the identification of vaccineresponders and the indication for a potential third vaccine dose. Strengths of our study is the prospective design and the ability to evaluate two patient populations under anti-CD20 therapy who have different underlying conditions that do both not greatly affect other axes of immune responses, such as in poly-immunosuppressed patients, or those suffering from lymphoma or leukemia.

In summary, our study suggests that patients with anti-CD20 treatment are able to mount potent T cell responses to mRNA COVID-19 vaccines similar to immunocompetent controls. 

Associations of baseline use of biologic or targeted synthetic DMARDs with COVID-19 severity in rheumatoid arthritis: Results from the COVID-19 Global Rheumatology Alliance physician registry

Disease-Modifying Therapies and Coronavirus Disease 2019 Severity in Multiple Sclerosis

COVID-19 outcomes in patients with inflammatory rheumatic and musculoskeletal diseases treated with rituximab: a cohort study

Effect of ocrelizumab on vaccine responses in patients with multiple sclerosis: The VELOCE study

Immunization responses in rheumatoid arthritis patients treated with rituximab: results from a controlled clinical trial

Rituximab impairs immunoglobulin (Ig)M and IgG (subclass) responses after influenza vaccination in rheumatoid arthritis patients

COVID-19 in multiple sclerosis patients: susceptibility, severity risk factors and serological response

Impaired humoral and cellular immunity after SARS-CoV2 BNT162b2 (Tozinameran) prime-boost vaccination in kidney transplant recipients

Efficacy of the BNT162b2 mRNA COVID-19 vaccine in patients with chronic lymphocytic leukemia

Immunogenicity and safety of the BNT162b2 mRNA COVID-19 vaccine in adult patients with autoimmune inflammatory and in the general population: a multicentre study

Humoral immune response to COVID-19 mRNA vaccine in patients with multiple sclerosis treated with high-efficacy disease-modifying therapies

Two X-linked agammaglobulinemia patients develop pneumonia as COVID-19 manifestation but recover

Correlates of protection against SARS-CoV-2 in rhesus macaques

Broad and strong memory CD4(+) and CD8(+) T cells induced by SARS-CoV-2 in UK convalescent individuals following COVID-19

Head-to-Head Evaluation of Five Automated SARS-CoV-2 Serology Immunoassays in Various Prevalence Settings

Targets of T Cell Responses to SARS-CoV-2 Coronavirus in Humans with COVID-19 Disease and Unexposed Individuals

COVID-19 vaccine BNT162b1 elicits human antibody and T(H)1 T cell responses

An mRNA Vaccine against SARS-CoV-2 -Preliminary Report

BNT162b2 vaccine induces neutralizing antibodies and poly-specific T cells in humans

but not other antirheumatic therapies, is associated with impaired serological response to SARS-CoV-2 vaccination in patients with rheumatic diseases

First-dose mRNA vaccination is sufficient to reactivate immunological memory to SARS-CoV-2 in subjects who have recovered from COVID-19

Ocrelizumab Depletes CD20⁺ T Cells in Multiple Sclerosis Patients

B cells reappear less mature and more activated after their anti-CD20-mediated depletion in multiple sclerosis

Humoral and T-cell responses to SARS-CoV-2 vaccination in patients receiving immunosuppression

Robust T Cell Immunity in Convalescent Individuals with Asymptomatic or Mild COVID-19

Anti-CD20 B-cell depletion enhances monocyte reactivity in neuroimmunological disorders

CD8(+) T cells contribute to survival in patients with COVID-19 and hematologic cancer

Rapid and stable mobilization of CD8+ T cells by SARS-CoV-2 mRNA vaccine

Efficacy and Safety of the mRNA-1273

SARS-CoV-2 Vaccine

Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine

BNT162b2 vaccine induces neutralizing antibodies and poly-specific T cells in humans

A c c e p t e d M a n u s c r i p t 14 A c c e p t e d M a n u s c r i p t 16