The impact of B cell deficiency on the humoral and cellular responses to SARS-CoV2 mRNA vaccination remains a challenging and significant clinical management question. We evaluated vaccine-elicited serological and cellular responses in 1) healthy individuals who were pre-exposed to SARS-CoV-2 (n = 21), 2) healthy individuals who received a homologous booster (mRNA, n = 19; or Novavax, n = 19), and 3) persons with multiple sclerosis on B cell depletion therapy (MS-αCD20) receiving mRNA homologous boosting (n = 36). Pre-exposure increased humoral and CD4 T cellular responses in immunocompetent individuals. Novavax homologous boosting induced a significantly more robust serological response than mRNA boosting. MS-α CD20 had an intact IgA mucosal response and an enhanced CD8 T cell response to mRNA boosting compared with immunocompetent individuals. This enhanced cellular response was characterized by the expansion of only effector, not memory, T cells. The enhancement of CD8 T cells in the setting of B cell depletion suggests a regulatory mechanism between B and CD8 T cell vaccine responses.

At the end of 2019, a highly transmissible, pathogenic, and novel coronavirus emerged, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), causing the coronavirus disease 2019 (COVID-19) pandemic. Globally, as of 5:57 pm, October 25, 2023, there have been 771,549,718 confirmed cases of COVID-19, including 6,974,473 deaths reported to the World Health Organization (https://covid19.who.int). Given the mild/asymptomatic nature of many infections, the true number of people with at least one infection is likely much higher. First-generation lipid nanoparticle mRNA vaccines, including Comirnaty (14) (Pfizer-BioNTech, previously BNT162b2) and Spikevax (5, 6) (Moderna, previously mRNA-1273), became available in the United States in December 2020 and remain the most used vaccines in many parts of the world (7). A viral vector-based vaccine (Janssen/J&J Ad26.COV2.S) was internationally assessed and approved in 2021 (8). The recombinant protein-based adjuvanted vaccine Novavax NVX-CoV2373 completed successful phase 3 efficacy clinical trials in the United States, Mexico, and the United Kingdom (9, 10) in 2021 and was approved for use in several different countries in 2022 (11). The current vaccines have proven to be invaluable tools for protecting public health and have saved countless lives. Many studies focused on their ability to induce serological and cellular immune responses have shown that they can induce short- and long-term (memory) immunity and protection. However, how this vaccine-elicited immunity (and protection) boosting frequency and regimen (i.e., following one brand or crossing brands) is affected by host immunocompetence remains unclear and warrants ongoing study.

Several studies have shown the effectiveness of vaccines with respect to symptomatic and asymptomatic SARS-CoV-2 infection, COVID-19 disease severity, and secondary transmission (12, 13). Additional vaccine boosters given months after initial vaccination have been shown to provide partial protection against novel variants, including Omicron (14, 15). However, the most protective immune responses are seen after a combination of vaccination and natural infection, also called hybrid immunity (1619). Bates et al. (20) showed that neutralizing Ab responses were more robust among those with hybrid immunity, including against the BA.2 variant. These titers significantly improved for those with longer vaccine-infection intervals, suggesting that longer interval periods in between vaccine boosting or infection/vaccination might lead to a more potent immunological response. Additionally, Hall et al. (21) demonstrated that infection-acquired immunity boosted with vaccination remained high more than 1 y postinfection, supporting that the combination of infection and vaccine-elicited immunity not only increases the magnitude of humoral immunity but also the durability. However, how these vaccine boosting regimen– and infection-elicited humoral and T cellular immunological responses are affected in immunocompromised individuals is not clear.

Head-to-head comparisons of SARS-CoV-2-spike-specific immune responses to Moderna mRNA-1273, Pfizer/BioNTech BNT162b2, Janssen Ad26.COV2.S, and Novavax NVX-CoV2373 examined in healthy individuals longitudinally for 6 mo demonstrated that all individuals made memory CD4+ T cells (22). Those who received mRNA or NVX-CoV2373 vaccinations had increased representation of circulating T follicular helper T cells and cytotoxic CD4 T cells (22). mRNA and Ad26.COV2.S vaccinees induced comparable Ag-specific CD8+ T cell frequencies, although only detectable in 60–67% of subjects at 6 mo (22). mRNA vaccinees had a substantial decline in Ab titers, whereas memory T and B cells were comparatively stable (22). Belik et al. (23) further showed that in all cases of mRNA and viral vector vaccines, third (booster) dose vaccine administration induced cross-neutralizing against multiple variants, including the Omicron BA.2 variant. Maringer et al. (24) demonstrated that spike-specific T cell responses were not further enhanced by booster vaccination in mRNA, vector vaccine, and heterologous (mRNA and viral vector mix) vaccination cohorts. These studies demonstrate that although infection- and/or vaccine-induced serological response wanes over time and increases with boosting, T and B cellular responses may remain more stable, regardless of mRNA or viral vector vaccine platform or boosting practices. However, the effect of third dose boosting compared between homologous (three doses of the same vaccine) mRNA (Pfizer and Moderna) and adjuvanted protein (Novavax) regimens has not been described.

All aforementioned studies were carried out in cohorts of immunocompetent individuals. Given that robust evidence for the presence of SARS-CoV-2 Ab responses as an indicator of immunity and protection, patients who are either 1) born with B cell defects or 2) receive treatments to deplete B cells are certain to generate decreased humoral SARS-CoV-2 vaccine/infection response, with the potential for increased susceptibility to severe COVID-19 disease. Anti-CD20 mAbs, commonly used for B cell malignancies, autoimmune conditions, and post-transplant immunosuppression, are particularly strong inhibitors for de novo vaccine responses. Indeed, SARS-CoV-2 vaccine seroconversion rates are around 10% or less in individuals vaccinated within 6 mo of receiving an anti-CD20 mAb (2527).

The impact of B cell deficiency on vaccine-elicited cellular (CD4 and/or CD8) responses is far more varied. Reduced CD8 T cell responses in patients with rheumatological disease receiving rituximab, an anti-CD20 mAb, have been observed in studies of influenza vaccines (2830) and tetanus toxoid vaccines (31). However, studies of SARS-CoV-2 vaccines have yielded more mixed data. A review of all COVID vaccine–elicited immunity studies performed on patients with X-linked agammaglobulinemia, who are born without any peripheral B cells, demonstrated that they all generated Ag-specific T cell responses (32). Similarly, vaccine responses in lymphoma patients or persons with multiple sclerosis on B cell depletion therapy (MS-αCD20s) showed similar frequencies in CD4 (33) and CD8 (33, 34) T cell responses, despite marked reduction in serological response. However, fewer patients with rheumatoid arthritis developed T cell responses regardless of their rituximab treatment status (35), and seroconverting and non-seroconverting MS-αCD20s had a trend toward lower T cell responses than healthy control subjects, but this was not significant, and there was again no difference based on seroconversion (36). However, in yet another study, although anti-CD20 therapy compromised circulating follicular helper T cell responses in MS-αCD20, type 1 helper T (Th1) cell priming was preserved, whereas CD8 T cell induction was augmented (33). Similarly augmented cellular responses from MS-αCD20 were observed in higher polyfunctional IFN-γ+ and IL-2+ T cell responses and strong T cell proliferation capacity compared with healthy controls (37). Furthermore, although there was no stratification performed based on the efficiency of B cell depletion, the authors found no difference in the T cell response observed between relapsing remitting or progressive MS, anti-CD20 therapy (rituximab versus ocrelizumab), or type of mRNA-based vaccine received (mRNA-1273 versus BNT162b2) (37). Altogether, these data suggest that the T cell response to COVID vaccination can be altered by the lack (or very reduced number) of peripheral B cells, regardless of the etiology. To clarify the role for B cell–depleting therapies, careful consideration must be given to whether B cells are effectively depleted and whether the underlying disease condition itself influences vaccine T cell responses, which could perhaps confound the interpretation of the influence of B cell–depleting therapies. What factors dictate these different results in cellular immune outcomes and what characterizes the augmented T cell responses elicited under such conditions are unknown and may affect treatment decisions and vaccination recommendations in these patients.

To address some of the knowledge gaps noted above, we set out to study SARS-CoV-2 mRNA and adjuvanted protein vaccine–elicited serological and cellular responses in 1) healthy individuals who were pre-exposed to SARS-CoV-2 (infection or vaccination) prior to the initial mRNA vaccine series, 2) healthy individuals who received three doses of the same vaccine, and 3) MS-αCD20s who underwent mRNA vaccine boosting.

Blood (n = 82) and nasal swabs (n = 49) were collected from healthy adults, ages 25–80 y old, enrolled and consented into study from Children’s Hospital Colorado and secondary locations at University of Colorado Denver, Anschutz Medical Campus, under an institutional review board–approved protocol (approval no. 21-3044) for which Dr. Hsieh is the principal investigator. Blood (n = 51) and nasal swabs (n = 38) were collected from individuals with multiple sclerosis (MS), ages 29–72 y old, enrolled and consented into study from the Neuroimmunology and Autoimmune Neurology clinics at the University of Colorado and from Children’s Hospital Colorado (approval nos. 18-1361 and 21-4092) for which Dr. Piquet is the principal investigator. Individuals were recruited between April 2021 and April 2022. Most participants received mRNA vaccination (Pfizer or Moderna); six individuals initially received single doses of vector-based vaccines under trial at that time (Astra Zeneca or Jansen) and then switched to a two-dose mRNA vaccine series (heterologous subjects in the pre-exposed group). All the patients with MS were actively receiving anti-CD20 therapy.

T cell response by activation-induced markers

PBMCs were thawed in 10 ml of warm 1× RPMI 1640 (Corning, catalog no. 15-040-CV) supplemented with 10% FBS (PEAK serum, catalog no. PS-FB3), 2 mM l-glutamine, 100 U/ml penicillin, and 100 mg/ml streptomycin (Life Technologies, catalog no. 10378016) (R10) and washed once in R10. For each condition, wells containing 2.5 × 106 cells in 200 ml were plated in 96-well round-bottom plates and rested overnight in a humidified incubator at 37°C, 5% CO2. After the overnight rest, the cells were stimulated for 24 h with costimulation 1ug/ml of anti-human CD28/CD49d (BD Biosciences, catalog no. 347690) and peptide receptor-binding domain (RBD) pools at a final concentration of 2 μg/ml. The peptide pools were designed to cover the SARS-CoV-2 spike protein receptor-binding domain (GenBank accession no. MT380724.1) derived from isolate Wuhan-Hu-1 (15-mer peptides overlapping by 11 amino acids). Unstimulated samples for each donor at each time point were treated with costimulation alone and costimulation plus 5 ug/ml PHA (Millipore Sigma, catalog no. 431784) as a positive control for PBMC viability and functional response to T cell stimulation. At 20 h after initiating stimulation, 3 μg/ml of brefeldin A adipogen (catalog no. AG-CN2-0018_M010) was added for intracellular cytokines detection. At 4 h after brefeldin A addition, the cells were washed in PBS (Life Technologies, catalog no. 14190-144) supplemented with 3% FBS (FACS buffer) for downstream flow cytometry analyses.

Flow cytometry acquisition and analysis

The cells were stained for 10 min at room temperature with Ghost dye Red 780 (Tonbo Biosciences, catalog no. 13-0865-T100) and washed once in FACS buffer + 3% FBS. Surface staining with Abs directed against FITC-conjugated anti-CD4 (RPA-T4, BioLegend, catalog no. 300506), PECy-7–conjugated anti-CD8 (RPA-T8, Tonbo Biosciences, catalog no. 60-0088-T100), Brilliant Violet 510–conjugated anti-CD45RA (HI100, BioLegend, catalog no. 304142), Alexa Fluor 700–conjugated anti-CD3 (SK7, BioLegend, catalog no. 344822), Brilliant Violet 421–conjugated anti-CD69 (FN50, BioLegend, catalog no. 310930), allophycocyanin-conjugated anti-CD137 (4B4-1, BioLegend, catalog no. 309810), PerCP-conjugated anti-CD14 (M5E2, BioLegend, 301848), PerCP-conjugated anti-CD19 (HIB19, BioLegend, catalog no. 302228), and Brilliant Violet 711–conjugated anti-CCR7 (G043H7, BioLegend, catalog no. 353227) was performed at 37°C for 15 min in FACS buffer. The cells were washed once in FACS buffer, fixed and permeabilized for 30 min at room temperature (eBioscience Foxp3/transcription factor staining buffer set, catalog no. 00-5523-00), and washed once in 1× permeabilization buffer prior to staining for intracellular phycoerythrin-conjugated anti–IFN-γ (4S.B3, BioLegend, catalog no. 502509) and Brilliant Violet 605–conjugated anti–TNF-α (MAb11, BioLegend, catalog no. 502936) overnight at 4°C. The cells were then washed once in 1× permeabilization buffer and resuspended in FACS wash prior to data acquisition. The data were acquired on a five-laser Beckman Coulter CytoFLEX S and were analyzed with CellEngine software by CellCarta. All data from activation-induced marker (AIM) expression assays were background-subtracted using paired unstimulated control samples. For memory T cell and helper T cell subsets, the AIM+ background frequency of non-naive T cells was subtracted independently for each subset. AIM+ cells were identified from non-naive T cell populations. AIM+ CD4 and CD8 T cells were defined by dual expression of CD69 and CD137. From AIM+ CD4 and CD8 T cells, intracellular IFN-γ production was evaluated, the population of interest was defined as IFN-γ+TNF-α. The lymphocytes were first gated to consider only live cells; then the lymphocytes were gated based on SSC-A and FSC-A and then CD3+ CD19CD14, cells expressing CD4 and CD8a were gated, and non-naive T cells (CD4 and CD8) were taken based on CCR7 and CD45RA markers to finally get to CD69+CD137+ AIM population of interest.

Multiplexed microsphere immunoassay

Plasma and nasal swabs from study subjects were tested for IgG and IgA detection against SARS-CoV-2. A multiplexed assay was developed by Dr. Ross Kedl (38) for the detection of anti–SARS-CoV-2 Abs to simultaneously quantify IgG and IgA against the spike RBD and N protein of SARS-COV-2. Additionally, specific seasonal coronavirus strains OC43, 229, and HKU were also included. Built-in negative control (BSA-conjugated beads) and positive control (tetanus toxoid) were added to the assay. Multiplex bead protein conjugation, sample preparation, sample incubation, and flow cytometry analysis were performed as described by Kedl et al. (38) and Schultz et al. (39). The geometric median fluorescence intensity (gMFI) of the anti-IgG specific for each protein/bead was computed for each sample using FlowJo software (version 10.7.1).

Enzyme-linked immunosorbent spot

The frequency of anti-RBD IgG+ B cells in the PBMC samples was determined using an enzyme-linked immunosorbent spot (ELISpot) assay. A total of 3 × 106 PBMC (per subject) were stimulated with ODN2395 (Invivogen, catalog no. tlrl-2395, 1–5 μM), IL-21 (GenScript, catalog no. P50061606, 10–100 ng/ml), IL-4 (GenScript, catalog no. B60071805, 10 ng/ml), CD40 (G28.5, BioXCell, catalog no. BE0189, 10 ng/ml), and B-Poly-S (MabTech, catalog no. CTL-HBPOLYS,) for 4.5 d at 37°C. Stimulated cells were added to the precoated ELISpot plates (Human IgG SARS-CoV-2 RBD ELISpot kit, Mab Tech, catalog no. 3850-4APW-R1-1), and the plates were incubated for 16–18 h at 37°C, following which the assay was developed to detect RBD-specific IgG. RBD-specific IgG detection was performed with an ELISpot and FluoroSpot reader (Advanced Imaging Devices GmbH, Strassberg, Germany) and Advanced Imaging Devices ELISpot software 7.0 for ELISpot counts.

Serological data preprocessing to minimize batch effects

The raw serological data underwent the following preanalytical steps: 1) data sorting per batch; 2) prebatch normalization data removal to maintain consistency with the remaining batches; 3) data extraction into a combined matrix; and 4) batch effect removal with the “removeBatchEffect()” function from the limma package (40), using the batch variable.

Hypothesis testing

Statistical comparisons among the groups were performed using the Kruskal–Wallis test (41), a nonparametric test designed for assessing differences between two or more independent groups. The Kruskal–Wallis test was executed using R (42), a statistical computing environment. Specifically, the kruskal.test() function in R was applied to compare serological and AIM measurements between the distinct groups. The single factor (B cell frequency) in Figs. 4C and 5 (B, C, E) was tested by applying the Mann–Whitney U test in Prism. A significance level of α = 0.05 was chosen to assess statistical significance for all tests.

Flow cytometry raw data tables (cell type frequencies for AIM assay, gMFI for the multiplexed microsphere immunoassay) have been deposited at Mendeley (https://data.mendeley.com, accession no. doi:10.17632/nz2jbfsddt.1) and are publicly available as of the date of publication. Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

We tested whether pre-exposure to SARS-CoV-2 Ags affects humoral and cellular immune responses to the mRNA primary vaccine series (Pfizer, two doses) in healthy individuals. We assessed 1) anti-RBD plasma and nasal IgG and IgA, 2) anti-RBD IgG+ B cell frequency (ELISpot), and 3) activation of SARS-CoV-2 Spike (S1) peptide pool-specific CD4 and CD8 non-naive T cells (AIMs CD69 and CD137) in PBMCs. All serological and cellular assays in this study refer to Ags derived from the ancestral Wuhan strain (WA1). Study subjects were considered pre-exposed to SARS-CoV-2 Ags if they had a record of either natural infection (confirmed by serology and PCR positivity; n = 15 subjects; a subset of subjects had repeated measures) or vaccination with a heterologous SARS-CoV-2 vaccine (e.g., Janssen or AstraZeneca; n = six subjects, a subset of subjects had repeated measures) prior to the primary mRNA vaccine series studied. Study subjects were considered unexposed (n = 54 subjects, a subset of subjects had repeated measures) if there was no history of infection symptoms and if nucleocapsid serology was negative at the time of sample collection (for demographic data see Table I). Sample time ranges were binned as prevaccination, 4–11 wk postvaccination, and 12–57 wk postvaccination (mean = 33.7; Supplemental Fig. 1 for sample time range details) with respect to their first vaccine dose for analysis.

Compared to unexposed subjects, pre-exposed subjects demonstrated increased levels of anti-RBD plasma IgG at all time points, with the greatest difference detected at the 4- to 11-wk time range (Fig. 1A, gMFI: unexposed 7.74e4 in blue, pre-exposed 10.5e4 in red, p = 0.011). To assess the dynamic range of humoral response to vaccination, we also compared matched subject anti-RBD responses across time ranges and found that pre-exposed subjects demonstrated a significantly higher differential response over the prevaccination to the 4- to 11-wk interval (Fig. 1B, gMFI: unexposed 5.11e4, pre-exposed 10.1e4, p = 0.033), consistent with the observation that pre-exposure confers the capacity to mount a more dynamic increase in anti-RBD IgG upon subsequent vaccination (20, 43, 44). Supporting the specificity of our Ab response measurement platform, Ab levels to related endemic coronavirus variants (OC43 and HKU) did not change in response to SARS-CoV-2 vaccination (Supplemental Fig. 1). Additionally, the frequency of anti-RBD IgG+ as a proportion of all IgG+ PBMC was significantly higher for the pre-exposed group after vaccination (Fig. 1C, median 2.01% unexposed, 13.0% pre-exposed; p = 0.017). Upon vaccination, unexposed subjects predominantly increased plasma IgG levels against the mRNA vaccine encoded RBD Ag but not the spike S2 region Ag (Fig. 1D, left). In contrast, naturally infected subjects exhibited IgG against both RBD and S2 prior to vaccination, and mRNA vaccination induced increases in anti-RBD IgG without loss of anti-S2 activity (Fig. 1D, right). Similar to the IgG response, the pre-exposed group demonstrated significantly higher anti-RBD IgA response 1) at the 4- to 11-wk time range (Fig. 1E, gMFI: unexposed 2.08e4, pre-exposed 2.89e4, p = 0.012) and 2) a greater differential of the prevaccination to the 4- to 11-wk interval (Fig. 1F, gMFI: unexposed 8.30e3, pre-exposed 18.6e3, p = 0.022). These results demonstrate that pre-exposure to SARS-CoV-2 Ags confers enhanced RBD-specific humoral immunity. To evaluate whether pre-exposure affects anti-RBD Ab levels at mucosal tissue sites, we tested nasal secretions collected between 33 and 57 wk postvaccination. At the 33- to 57-wk time range, the pre-exposed subjects had similar nasal anti-RBD IgG compared with unexposed (Fig. 1G1, gMFI: unexposed 3.18e4, pre-exposed 3.73e4, p = 0.101) but significantly elevated nasal IgA (Fig. 1H, gMFI: unexposed 9.09e3, pre-exposed 42.1e3, p = 0.020), supporting the observation that these subjects also had a more dynamic plasma IgA response from prevaccination to 4–11 wk and that this increased response translated into durable IgA levels at mucosal sites of action thereafter.

We evaluated the Ag-reactive T cell response to vaccination by measuring 1) the frequency of AIM+ non-naive CD4 and CD8 T cells in response to SARS-CoV-2 RBD peptide pools (subtracting unstimulated background and adding PHA or PHA as a stimulation-positive control) and 2) the frequency of those AIM+ cells expressing above-threshold (unstimulated) levels of IFN-γ+ (out of non-naive CD4 and CD8 T cells, respectively, and defining the IFN-γ+ population of interest as IFN-γ+TNF-α). This was applied uniformly to all individuals (Fig. 2A). Compared to unexposed, pre-exposed subjects had a higher frequency of CD4+AIM+ (Fig. 2B, median frequency: unexposed −0.077%, pre-exposed 0.001%; p = 0.023) and CD4+AIM+IFN-γ+ T cells (Fig. 2C, median frequency: unexposed 0.008%; pre-exposed 0.059%; p = 0.025) at the 4- to 11-wk time range. Prior to vaccination, whereas the pre-exposed individuals had frequency of CD4+AIM+ T cells similar to that of the unexposed (Fig. 2B, median frequency: unexposed −0.183%, pre-exposed −0.103%; p = 0.768), a higher frequency of those cells produced IFN-γ, supporting an enhanced antiviral response in the pre-exposed group (Fig. 2C, median frequency: unexposed −0.031%, pre-exposed −0.004%; p = 6.5e-3).

Pre-exposed subjects also demonstrated a significant increase in the frequencies of both CD4+AIM+ and CD4+AIM+IFN-γ+ T cells from the prevaccination to 4- to 11-wk time range (AIM+p = 5.0e−3; AIM+IFN-γ+p = 3.9e−3) and a decrease in these populations from 4–11 wk to 12–57 wk postvaccination (AIM+p = 0.043; AIM+IFN-γ+p = 0.012), supporting the need for boosting to maintain T cell–mediated Ag-specific immunity. Unexposed subjects only showed this pattern of increase followed by decrease over time for the CD4+AIM+IFN-γ+ cells (prevaccination to 4–11 wk: p = 1.1e−3; 4–11 wk to 12–57 wk: p = 0.044). Surprisingly, given the well documented importance of CD8+ T cells for viral clearance, exposure-based differences in Ag-specific CD4 T cell responses were not matched by any significant exposure-based differences in CD8 T cells (Fig. 2D, 2E).

Taken together, these results support the established (20, 43, 44) idea that pre-exposure to SARS-CoV-2 Ags, whether by natural infection or heterologous vaccination (e.g., Janssen or AstraZeneca), confers a more rapid and robust response to the mRNA primary vaccine series with the following adaptive immunological features: 1) increased circulating anti-RBD Abs (Fig. 1A, 1D); 2) increased frequency of anti-RBD IgG-producing B cells (Fig. 1C); 3) more dynamic increases in plasma anti-RBD IgG and IgA (Fig. 1B, 1E), with 4) sustained presence of mucosal-acting IgA at mucosal tissues months after vaccination (Fig. 1F–G); and 5) increased frequency of Ag-reactive (AIM+) and IFN-γ producing CD4 T cells at 8 wk following vaccination (Fig. 2A–B).

Having demonstrated that humoral and cellular mRNA vaccine–elicited immune responses are enhanced by pre-exposure to SARS-CoV-2 Ags and that such responses wane in the 12- to 57-wk postvaccination time range, we tested whether a single homologous booster dose approximately 6 mo after the primary series elicited significant changes in these vaccine-elicited immune responses. We evaluated subjects who received 1) the two widely used mRNA vaccines (Pfizer, n = 7, and Moderna, n = 12) and 2) the Novavax protein subunit vaccine (n = 19) at approximately 4–6 mo after the second dose of primary series (pre) and approximately 4 wk after boost (post).

Consistent with its documented ability to elicit higher levels of neutralizing Ab in primary vaccination (45, 46), Novavax induced a more robust postboost anti-RBD plasma IgG increase than the two mRNA vaccines (mRNA median fold change = 1.91, Moderna in teal and Pfizer in gray; Novavax median fold change = 6.39 in orange; p value = 0.019; Fig. 3A). However, the booster-induced anti-RBD plasma IgA increase was similar between the mRNA vaccines and Novavax (mRNA median fold change = 1.39; Novavax median fold change = 0.071; p value = 0.31; Fig. 3B). The booster-induced increase in anti-RBD plasma IgG (1.91-fold) and IgA (1.39-fold) was similar between the two mRNA vaccines (p values = 0.80 and 0.13 for anti-IgG and -IgA, respectively), and hence they were grouped together for comparison against Novavax, an adjuvanted protein vaccine modality. mRNA vaccine recipients in our study cohort received their booster doses between 33 and 57 wk after the second dose of their primary series, whereas Novavax recipients all received their boosts between 34 and 36 wk after their second dose. However, any differences in the anti-RBD IgG/A response between vaccine brands is not attributable to the difference in this time interval because mRNA recipients showed no statistical difference in a test between early (<37 wk) and late (>37 wk) booster timing (Supplemental Fig. 2A, 2B).

Abs to related endemic coronavirus variants (OC43 and HKU) did not increase in response to boosting with any vaccine brand included in this study, supporting the specificity of the response (Supplemental Fig. 2C–F). Taken together, these results indicate that boosting with any of the three vaccines induced increases in systemic IgG specific to the RBD Ag. In particular, the Novavax protein subunit boost exhibited higher levels of systemic IgG. SARS-CoV-2 S1-specific CD4 and CD8 T cellular responses across different boosting modalities all resulted in increases postvaccination (Supplemental Fig. 3), although they were statistically insignificant. Although these findings are in line with previously published work (24), the heterogeneous preboosting biospecimen sampling timeframe likely resulted in high variance in preboost cellular AIM values and challenges in statistical significance.

SARS-CoV-2 boosting vaccination recommendation and practices are modified for patients who are immunosuppressed, because they are at increased risk of severe infection complications. Because vaccine effectiveness data (i.e., protection) has been shown based on circulating Ab responses, patients with humoral defects caused by primary genetic etiologies or secondary immunomodulation do not have clear recommendations as to how many booster doses they should receive and with what interval to experience protection. Although T cell immune response may be intact for some of these patients, humoral and cellular SARS-CoV-2 vaccine–elicited responses from peripheral circulating and mucosal sites have not been fully assessed in response to one booster vaccine. We tested whether B cell–depleted patients with MS (MS-α CD20) and healthy immunocompetent individuals showed differences in vaccine-elicited immune responses to a single booster dose of mRNA vaccine administered 5–10 mo (mean = 7.3 mo) after the primary series. We evaluated plasma and nasal anti-RBD IgG/A, and SARS-CoV-2 S1-specific T cell responses. Sample time points were binned as preboost (approximately 4–6 mo from the second dose of primary series), 4–7 wk postboost, and 8–26 wk postboost (mean = 15 mo; Supplemental Fig. 4 for sample time range details) for analysis.

The results indicate that MS-α CD20s had lower levels of plasma anti-RBD IgG compared with healthy controls at all time points tested (Fig. 4A, gMFI: preboost healthy control [HC] 3.177e4, preboost MS 3.166e4, p = 1.3e−3; 4–7 wk HC 6.094e4, 4–7 wk MS 3.166e4, p = 2.7e−4; 8–26 wk HC 7.059e4, 8–26 wk MS 1.191e4, p = 0.039). Unlike healthy controls, MS-α CD20s showed no significant increase in anti-RBD IgG across time points (Fig. 4A, HC preboost versus 4–7 wk, p = 1.7e−4, preboost versus 8–26 wk, p = 0.049). In contrast, MS-α CD20s do have mucosal anti-RBD IgA levels comparable to healthy controls at all time points (Fig. 4B, gMFI: preboost HC 6.502e3, preboost MS 8.691e3, p = 0.537; 4–7 wk HC 3.490e4, 4–7 wk MS 1.429e4, p = 0.187; 8–26 wk HC 3.514e4, 8–26 wk MS 1.335e4, p = 0.089). These data indicate that MS-α CD20s do not increase their anti-RBD plasma IgG after boosting as healthy controls do, although they may maintain a source of mucosal anti-RBD IgA despite their B cell depletion (Fig. 4B). Comparison of 4- to 7-wk anti-RBD IgG levels showed that individuals in the MS-α CD20 group with fewer than 1% B cells (n = 20) have significantly lower levels than healthy controls (n = 11), whereas ostensibly B cell–depleted subjects with greater than 1% B cells (n = 7) have levels comparable to healthy controls (Fig. 4C, gMFI: HC 7.06e4, MS > 1% 5.50e4, MS < 1% 1.38e4, p < 0.001 for HC versus MS < 1%). We showed the healthy controls’ response to boost to be SARS-CoV-2 RBD-specific by testing Ags from seasonal variant coronaviruses (Supplemental Fig. 4A–D). These data indicate that the lack of anti-RBD Abs is a downstream effect of B cell depletion with anti-CD20 therapy and that a minimal threshold of B cell frequency seems to be required for Ab production peripherally and at the mucosal sites, with a proportional Ab production response above such threshold.

We evaluated the Ag-reactive T cell response to vaccination by measuring the frequency of non-naive CD8+AIM+ T cells in response to SARS-CoV-2 RBD total S1 peptide pools. MS-α CD20s had frequencies of CD8+AIM+ T cells and CD8+AIM+IFN-γ+ (from CD8+ non-naive T) comparable to healthy controls at all time ranges tested (Fig. 5A, AIM+ median frequency: preboost HC 0.491, preboost MS 0.709, p = 0.972; 4–7 wk HC 0.879, 4–7 wk MS 1.829, p = 0.281; 8–26 wk HC 0.505, 8–26 wk MS 1.527, p = 0.138; Fig. 5D, AIM+IFN-γ+ median frequency: preboost HC 0.040, preboost MS 0.011, p = 0.068; 4–7 wk HC 0.035, 4–7 wk MS 0.179, p = 0.51; 8–26 wk HC 0.023, 8–26 wk MS 0.124, p = 0.32). They also show significant increases in the frequency of these cells (Fig. 5A, preboost versus 4–7 wk, p = 0.048, preboost versus 8–26 wk, p = 0.049) and the frequency of these cells that make IFN-γ (Fig. 5D, preboost versus 4–7 wk, p = 0.024, preboost versus 8–26 wk, p = 0.036), across pre- and postboosting time ranges, whereas healthy controls do not. Remarkably, the frequency of both booster-induced CD8+AIM+ and CD8+AIM+IFN-γ+ T cells at all postboost time ranges is significantly greater for MS-α CD20 with fewer than 1% B cells when compared with either healthy controls or patients with MS with greater than 1% B cells (Fig. 5B, AIM+ median frequency: HC 0.62, MS > 1% 0.64, MS < 1% 3.0, p = 0.003 for HC versus MS < 1%; Fig. 5E, AIM+IFN-γ+ median frequency: HC 0.024, MS > 1% 0.0053, MS < 1% 0.25, p = 0.02 for HC versus MS < 1%).

Gating the AIM+ cells from Fig. 5B by CCR7 and CD45RA to define central memory (TCM, CCR7+CD45RA), effector memory (TEM, CCR7CD45RA), and effector memory expressing CD45RA (TEMRA, CCR7CD45RA+) T cell subsets demonstrated that the elevated AIM+ frequency observed for the MS < 1% group was completely attributable to a significantly higher frequency of TEM and TEMRA cells compared with HC, with no difference in TCM frequency (Fig. 5C, TEM median frequency: HC 0.280, MS < 1% 1.47, p = 0.0133; TEMRA median frequency: HC 0.153, MS < 1% 0.778, p = 0.0060). The frequency of TEM cells for the MS < 1% group also trended higher than the MS > 1% group, although the p value was greater than 0.05 (Fig. 5C, TEM median frequency: MS > 1% 0.293, MS < 1% 1.47, p = 0.0613). Notably, evaluation of the frequency of CD8+AIM+IFN-γ+ T cells for each individual B cell–depleted patient across time points as a percentage of their maximal response demonstrated that nearly all (9 of 11) mount their maximal response approximately 6–8 weeks postboosting and decline around 12 wk postboosting (Fig. 5F). However, such uniformity is not observed when evaluating the frequency of CD8+AIM+ in this manner, suggesting that the cytokine-producing effector T cell memory response follows a predictable crescendo/decrescendo trajectory (Supplemental Fig. 4E). These data support the notion that although MS-α CD20s have a deficient systemic Ab response, they have the capacity to increase Ag-specific CD8 T effector cells in response to a booster dose of mRNA vaccine. These findings suggest that there is either 1) a compensatory response of cellular immunity in the setting of impaired humoral immunity or 2) a regulatory process that vaccine-elicited B cell responses may have on CD8 T cells.

In this study, we evaluated SARS-CoV-2 vaccine–elicited humoral and cellular responses with respect to three different variables: 1) antigenic exposure prior to mRNA vaccination, 2) different vaccine boosting schemes, and 3) vaccine booster response in immunocompetence and B cell depletion in the context of MS. Most notably, our findings that B cell–depleted patients exhibit stable mucosal anti-RBD IgA production, together with expanded Ag-specific CD8 T cells composed primarily of effector cells, present important considerations for future work aimed at optimizing adaptive immune responses in the setting of humoral immunosuppression.

An increasing number of studies in rheumatologic diseases and in MS indicate preserved but altered T cell responses to COVID-19 vaccination in anti-CD20–treated patients (33, 36, 4751). Specifically, it was shown that MS-α CD20s who failed to develop circulating Abs following vaccination were able to generate robust T cell responses compared with patients with preserved humoral vaccine responses (48). We observed that MS-α CD20s had mucosal anti-RBD IgA levels comparable to immunocompetent controls at all time points evaluated (Fig. 4B), despite a clear deficit in their serum IgG against this Ag. This implies that anti-CD20 therapy spares IgA-producing B cells, likely explained by absent CD20 surface expression on plasmablasts/plasma cells in mucosal tissues (52). Case reports have described persistent IgA production by patients on anti-CD20 therapy in the setting of rheumatologic conditions and SARS-CoV-2 infection (5254). Notably, Mei et al. (52) provided evidence that mucosal IgA+ plasmablasts escape deletion by rituximab in a cohort of rheumatoid arthritis patients. Our finding extends the observation of this phenomenon to the setting of vaccination and raises the question of how it could be leveraged to best support protection from infection in patients on humoral immunosuppression. Although T cell–mediated immunity has received due attention as the most significant correlate of protection against severe disease (55), mucosal neutralizing Abs against SARS-CoV-2 remain an important correlate of protection against viral entry, and strategies to promote their production continue to be viable goals in vaccine development, notably in the field of respiratory routes of vaccine delivery (56). It has been shown that salivary anti-SARS-CoV-2 IgA levels are not as long-lived as IgG levels, which derive primarily from passive diffusion from the serum at gingival tissues (57, 58). Thus, it is reasonable to consider that maintenance of mucosal IgA-mediated protection of individuals unable to produce systemic IgG could benefit from a regimen of repeat vaccine boosting. Our data demonstrate that mucosal immunity against viral entry in B cell–depleted patients could plausibly be reinforced by such a maintenance strategy. Future work should be done to determine whether booster doses indeed enhance viral neutralization capacity in the mucosa of these patients and how long this effect endures.

Our data demonstrate that enhanced T cell responses in the setting of impaired Ab production is recapitulated when booster doses are administered 5–10 mo after primary vaccination (Figs. 4 and 5). This implies that MS-αCD20 patients may benefit from a regimen of repeat boosting to maintain compensatory cell-mediated immunity in the face of deficient or absent Ab production. Although our study does not evaluate the effect on protection from disease directly, a growing body of evidence on the correlates of protection from COVID-19 disease emphasizes the central importance of Ag-specific CD8 T cells (55, 59, 60). We found significant enhancement in spike-specific CD8 T cells (Fig. 5A), as well as those producing IFN-γ within that subset (Fig. 5D), both 4 wk after boosting and at further time points (8–26 wk), indicating that a cytotoxic effector response can be induced expediently and that it may endure for months thereafter. However, we also note that the frequency of IFN-γ–positive cells, when analyzed as percentage of a subject’s maximal response, begins to fall after peaking at 4–6 wk postboost (Fig. 5F). Further studies are warranted to test the durability of the response at later time points than those examined in this study.

We also segregated MS-αCD20s according to their measured B cell frequency, demonstrating that the discordant relationship between Ab deficit and CD8 T cell enhancement is indeed dependent on low (<1%) B cell frequency (Figs. 4C and 5C). This shows that the observed increase in Ag-specific CD8 T cells is not generalizable to all MS-αCD20 in our study cohort but is specific to those whose B cells are effectively peripherally depleted (<1% peripheral B cells). Given the heterogeneity of depth of peripheral B cell depletion across patients, even when treated with the same anti-CD20 regimen, monitoring B cell frequency presents an important consideration in determining which patients could benefit most from a vaccine boosting recommendation and at what time relative to their anti-CD20 therapy. Furthermore, these findings imply that the underlying biology of CD8 T cell responses to vaccination depend on a regulatory interrelationship with B cells. Importantly, the enhanced CD8 T cell response observed in our study is attributable to increased spike-specific TEM and TEMRA but not TCM cells (Fig. 5C). One implication of these data is that compromised B cell–dependent memory CD8 T formation may still be a longer-term consideration in this patient group. Hence, long intervals between booster doses could potentially leave them susceptible to lose the amount of cell-mediated protection they have. Overall, it will be necessary to consider both positive and negative consequences of B cell deficiency (naturally occurring or therapeutically induced) in the generation of T cell responses to vaccines.

It has been well established that humoral immunity is higher in individuals who experienced SARS-CoV-2 natural infection before or after vaccination, showing a significant increase in Ab response (16, 61, 62). Consistent with these publications, our data demonstrated that compared with unexposed healthy individuals, pre-exposed (natural infection or heterologous vaccination) study participants showed an increased systemic and mucosal IgG and IgA response (Fig. 1). Upon vaccination, preinfected subjects exhibited increases in anti-RBD IgG without loss of existing anti-S2 activity, indicating that vaccination following infection can enhance reactivity against a broader repertoire of viral Ags than vaccination alone, which would have the potential for more cross-reactivity across viral strains (Fig. 1). SARS-CoV-2 homology with other viruses such as HKU1 and OC43, among others, has been described, suggesting that cross-reactive immune responses may play an important role in SARS-CoV-2 infection and/or vaccination outcomes (63). Serological response to HKU1 and OC43 did not change with SARS-CoV-2 vaccination in either pre-exposed or unexposed individuals, supporting the specificity of our serological measurements (Supplemental Fig. 1). Our work also demonstrates that the spike-specific cellular response elicited by CD4 T cells was higher in the pre-exposed group, specifically at 8 wk after primary vaccination (Fig. 2B–C). Although we were not able to detect a significant difference in the CD8 T cell responses between the pre-exposed and unexposed groups, the magnitude of the response was still higher in the pre-exposed group (Fig. 2D–E). It has been previously reported that the frequency of circulating SARS-CoV-2-specific CD8 T cells in subjects recovered from COVID infection peaked 1 mo postinfection (61). Hence, it is possible that our dataset missed such a maximal response given the lack of biospecimen collection at that time point.

Some studies have shown that heterologous two-dose vaccinations with a vector-based vaccine followed by an mRNA-based vaccine elicit higher levels of neutralizing Abs than homologous two-dose vaccinations with only vector-based or only mRNA-based vaccines (23). Our data support that our pre-exposed cohort (composed of heterologous vaccinees and previously COVID-infected people) indeed show an enhanced Ab response. Additionally, homologous booster vaccination with the protein subunit vaccine (Novavax) demonstrated a more robust systemic serological response than homologous boosting with mRNA vaccines (Fig. 3). A study in macaques demonstrated that both heterologous Novavax and homologous mRNA booster vaccines induced low frequencies of circulating spike-specific T cells, whereas the Novavax booster induced a greater proportion of Ag-specific plasmablasts and blunted viral replication in the upper airway (although both boosters blunted viral replication in the lower airway) (64). To our knowledge, our finding that Novavax boosting induced enhanced systemic IgG compared with mRNA boosting is the first in the context of a homologous boosting regimen with this vaccine, indicating that the effect is attributable to the platform and not a hybrid prime-boost regimen.

The authors have no financial conflicts of interest.

We thank the study participants and their families for their contribution in the study. We also thank the Children’s Hospital Colorado Research Institute and University of Colorado Anschutz Medical Campus clinical coordinators, regulatory staff, and research project managers for the development of the institutional review board protocols and studies and the enrollment of study subjects. We thank the Department of Neurology, Section of Neuroimmunology and Department of Pediatrics, Section of Allergy and Immunology physicians and nurses, who also helped with the enrollment of study subjects. We thank the Human Immune Monitoring Shared Resource RRID:SCR_021985 for their expertise in ELISpot assays.

This work was supported by National Institutes of Health Grants R01 AI174303 (to R.M.B., B.C.-M., and E.W.Y.H.) and R01 AI148919, R01 AI126899, and R01 AI066121 (to R.M.K., J.K., M.G.M., and C.R.); the Boettcher Foundation COVID Biomedical Research Innovation Fund (to R.M.B., B.C.-M., and E.W.Y.H.); NIAMS, National Institutes of Health Grant P30 AR079369 (to K.D.D.); the William P. Arend Endowment (to K.D.D.); University of Colorado Cancer Center Grant P30CA046934 (to the Human Immune Monitoring Shared Resource); the Grohne-Stepp Endowment of the University of Colorado Cancer Center (to D.G. and T.G.); the Jeffrey Modell Foundation Immunodeficiency Center; the Boettcher Foundation COVID Biomedical Research Innovation Fund; Lyda Hill Philanthropy; the Rocky Mountain Multiple Sclerosis Center (to A.L.P.); the University of Colorado (to A.L.P.); and the Drake Family in the name of Susan Drake (to A.L.P.). The data generated was also supported by the University of Colorado Department of Immunology and Microbiology pilot funds and University of Colorado Anschutz Medical Campus enrichment funds.

The flow cytometry raw data tables presented in this article have been submitted to the Mendeley (https://data.mendeley.com) under accession number doi: 10.17632/nz2jbfsddt.1.

The online version of this article contains supplemental material.

AIM

activation-induced marker

ELISpot

enzyme-linked immunosorbent spot

gMFI

geometric median fluorescence intensity

HC

healthy control

MS

multiple sclerosis

MS-αCD20

patient with multiple sclerosis on B cell depletion therapy

RBD

receptor-binding domain

TCM

central memory subset of T cells

TEM

effector memory subset of T cells

TEMRA

effector memory expressing CD45RA subset of T cells

1
Hause
,
A. M.
,
P.
Marquez
,
B.
Zhang
,
T. R.
Myers
,
J.
Gee
,
J. R.
Su
,
P. G.
Blanc
,
A.
Thomas
,
D.
Thompson
,
T. T.
Shimabukuro
,
D. K.
Shay
.
2022
.
Safety monitoring of bivalent COVID-19 mRNA vaccine booster doses among persons aged ≥12 years – United States, August 31–October 23, 2022
.
MMWR. Morb. Mortal. Wkly. Rep.
71
:
1401
1406
.
2
Tartof
,
S. Y.
,
J. M.
Slezak
,
H.
Fischer
,
V.
Hong
,
B. K.
Ackerson
,
O. N.
Ranasinghe
,
T. B.
Frankland
,
O. A.
Ogun
,
J. M.
Zamparo
,
S.
Gray
, et al
2021
.
Effectiveness of mRNA BNT162b2 COVID-19 vaccine up to 6 months in a large integrated health system in the USA: a retrospective cohort study
.
Lancet
398
:
1407
1416
.
3
Vogel
,
A. B.
,
I.
Kanevsky
,
Y.
Che
,
K. A.
Swanson
,
A.
Muik
,
M.
Vormehr
,
L. M.
Kranz
,
K. C.
Walzer
,
S.
Hein
,
A.
Güler
, et al
2021
.
BNT162b vaccines protect rhesus macaques from SARS-CoV-2
.
Nature
592
:
283
289
.
4
Winokur
,
P.
,
J.
Gayed
,
D.
Fitz-Patrick
,
S. J.
Thomas
,
O.
Diya
,
S.
Lockhart
,
X.
Xu
,
Y.
Zhang
,
V.
Bangad
,
H. I.
Schwartz
, et al
C4591031 Clinical Trial Group
.
2023
.
Bivalent omicron BA.1-adapted BNT162b2 booster in adults older than 55 years
.
N. Engl. J. Med.
388
:
214
227
.
5
Andrews
,
N.
,
J.
Stowe
,
F.
Kirsebom
,
S.
Toffa
,
T.
Rickeard
,
E.
Gallagher
,
C.
Gower
,
M.
Kall
,
N.
Groves
,
A. M.
O’Connell
, et al
2022
.
COVID-19 vaccine effectiveness against the omicron (B.1.1.529) variant
.
N. Engl. J. Med.
386
:
1532
1546
.
6
Jackson
,
L. A.
,
E. J.
Anderson
,
N. G.
Rouphael
,
P. C.
Roberts
,
M.
Makhene
,
R. N.
Coler
,
M. P.
McCullough
,
J. D.
Chappell
,
M. R.
Denison
,
L. J.
Stevens
, et al;
mRNA-1273 Study Group
.
2020
.
An mRNA vaccine against SARS-CoV-2—preliminary report
.
N. Engl. J. Med.
383
:
1920
1931
.
7
Mathieu
,
H. R. E
,
L.
Rodés-Guirao
,
C.
Appel
,
C.
Giattino
,
J.
Hasell
,
B.
Macdonald
,
S.
Dattani
,
D.
Beltekian
,
E.
Ortiz-Ospina
,
M.
Roser
.
2020
.
Coronavirus pandemic (COVID-19)
. Available at: https://ourworldindata.org/covid-vaccinations.
8
Sadoff
,
J.
,
G.
Gray
,
A.
Vandebosch
,
V.
Cárdenas
,
G.
Shukarev
,
B.
Grinsztejn
,
P. A.
Goepfert
,
C.
Truyers
,
H.
Fennema
,
B.
Spiessens
, et al
ENSEMBLE Study Group
.
2021
.
Safety and efficacy of single-dose Ad26.COV2.S vaccine against COVID-19
.
N. Engl. J. Med.
384
:
2187
2201
.
9
Dunkle
,
L. M.
,
K. L.
Kotloff
,
C. L.
Gay
,
G.
Áñ
ez
,
J. M.
Adelglass
,
A. Q.
Barrat Hern
á
ndez
,
W. L.
Harper
,
D. M.
Duncanson
,
M. A.
McArthur
,
D. F.
Florescu
, et al
2019nCoV-301 Study Group
.
2022
.
Efficacy and safety of NVX-CoV2373 in adults in the United States and Mexico
.
N. Engl. J. Med.
386
:
531
543
.
10
Heath
,
P. T.
,
E. P.
Galiza
,
D. N.
Baxter
,
M.
Boffito
,
D.
Browne
,
F.
Burns
,
D. R.
Chadwick
,
R.
Clark
,
C.
Cosgrove
,
J.
Galloway
, et al
2019nCoV-302 Study Group
.
2021
.
Safety and efficacy of NVX-CoV2373 COVID-19 vaccine
.
N. Engl. J. Med.
385
:
1172
1183
.
11
Parums
,
D. V.
2022
.
First approval of the protein-based adjuvanted nuvaxovid (NVX-CoV2373) novavax vaccine for SARS-CoV-2 could increase vaccine uptake and provide immune protection from viral variants
.
Med. Sci. Monit.
28
:
e936523
.
12
Hall
,
V. J.
,
S.
Foulkes
,
A.
Saei
,
N.
Andrews
,
B.
Oguti
,
A.
Charlett
,
E.
Wellington
,
J.
Stowe
,
N.
Gillson
,
A.
Atti
, et al
SIREN Study Group
.
2021
.
COVID-19 vaccine coverage in health-care workers in England and effectiveness of BNT162b2 mRNA vaccine against infection (SIREN): a prospective, multicentre, cohort study
.
Lancet
397
:
1725
1735
.
13
Lopez Bernal
,
J.
,
N.
Andrews
,
C.
Gower
,
C.
Robertson
,
J.
Stowe
,
E.
Tessier
,
R.
Simmons
,
S.
Cottrell
,
R.
Roberts
,
M.
O’Doherty
, et al
2021
.
Effectiveness of the Pfizer-BioNTech and Oxford-AstraZeneca vaccines on COVID-19 related symptoms, hospital admissions, and mortality in older adults in England: test negative case-control study
.
BMJ
373
:
n1088
.
14
Choi
,
A.
,
M.
Koch
,
K.
Wu
,
L.
Chu
,
L.
Ma
,
A.
Hill
,
N.
Nunna
,
W.
Huang
,
J.
Oestreicher
,
T.
Colpitts
, et al
2021
.
Safety and immunogenicity of SARS-CoV-2 variant mRNA vaccine boosters in healthy adults: an interim analysis
.
Nat. Med.
27
:
2025
2031
.
15
W. F.
Garcia-Beltran
.,
K. J.
St Denis
,
A.
Hoelzemer
,
E. C.
Lam
,
A. D.
Nitido
,
M. L.
Sheehan
,
C.
Berrios
,
O.
Ofoman
,
C. C.
Chang
,
B. M.
Hauser
, et al
2022
.
mRNA-based COVID-19 vaccine boosters induce neutralizing immunity against SARS-CoV-2 Omicron variant
.
Cell
185
:
457
466.e4
.
16
Bates
,
T. A.
,
S. K.
McBride
,
H. C.
Leier
,
G.
Guzman
,
Z. L.
Lyski
,
D.
Schoen
,
B.
Winders
,
J. Y.
Lee
,
D. X.
Lee
,
W. B.
Messer
, et al
2022
.
Vaccination before or after SARS-CoV-2 infection leads to robust humoral response and antibodies that effectively neutralize variants
.
Sci. Immunol.
7
:
eabn8014
.
17
Goldberg
,
Y.
,
M.
Mandel
,
Y. M.
Bar-On
,
O.
Bodenheimer
,
L. S.
Freedman
,
N.
Ash
,
S.
Alroy-Preis
,
A.
Huppert
,
R.
Milo
.
2022
.
Protection and waning of natural and hybrid immunity to SARS-CoV-2
.
N. Engl. J. Med.
386
:
2201
2212
.
18
Pilz
,
S.
,
V.
Theiler-Schwetz
,
C.
Trummer
,
R.
Krause
,
J. P. A.
Ioannidis
.
2022
.
SARS-CoV-2 reinfections: overview of efficacy and duration of natural and hybrid immunity
.
Environ. Res.
209
:
112911
.
19
Wratil
,
P. R.
,
M.
Stern
,
A.
Priller
,
A.
Willmann
,
G.
Almanzar
,
E.
Vogel
,
M.
Feuerherd
,
C. C.
Cheng
,
S.
Yazici
,
C.
Christa
, et al
2022
.
Three exposures to the spike protein of SARS-CoV-2 by either infection or vaccination elicit superior neutralizing immunity to all variants of concern
.
Nat. Med.
28
:
496
503
.
20
Bates
,
T. A.
,
H. C.
Leier
,
S. K.
McBride
,
D.
Schoen
,
Z. L.
Lyski
,
D. X.
Lee
,
W. B.
Messer
,
M. E.
Curlin
,
F. G.
Tafesse
.
2023
.
An extended interval between vaccination and infection enhances hybrid immunity against SARS-CoV-2 variants
.
JCI Insight
8
:
e165265
.
21
Hall
,
V.
,
S.
Foulkes
,
F.
Insalata
,
P.
Kirwan
,
A.
Saei
,
A.
Atti
,
E.
Wellington
,
J.
Khawam
,
K.
Munro
,
M.
Cole
, et al
SIREN Study Group
.
2022
.
Protection against SARS-CoV-2 after COVID-19 vaccination and previous infection
.
N. Engl. J. Med.
386
:
1207
1220
.
22
Zhang
,
Z.
,
J.
Mateus
,
C. H.
Coelho
,
J. M.
Dan
,
C. R.
Moderbacher
,
R. I.
Galvez
,
F. H.
Cortes
,
A.
Grifoni
,
A.
Tarke
,
J.
Chang
, et al
2022
.
Humoral and cellular immune memory to four COVID-19 vaccines
.
Cell
185
:
2434
2451.e17
.
23
Belik
,
M.
,
P.
Jalkanen
,
R.
Lundberg
,
A.
Reinholm
,
L.
Laine
,
E.
Väisänen
,
M.
Skön
,
P. A.
Tähtinen
,
L.
Ivaska
,
S. H.
Pakkanen
, et al
2022
.
Comparative analysis of COVID-19 vaccine responses and third booster dose-induced neutralizing antibodies against Delta and Omicron variants
.
Nat. Commun.
13
:
2476
.
24
Maringer
,
Y.
,
A.
Nelde
,
S. M.
Schroeder
,
J.
Schuhmacher
,
S.
Hörber
,
A.
Peter
,
J.
Karbach
,
E.
Jäger
,
J. S.
Walz
.
2022
.
Durable spike-specific T cell responses after different COVID-19 vaccination regimens are not further enhanced by booster vaccination
.
Sci. Immunol.
7
:
eadd3899
.
25
Gurion
,
R.
,
U.
Rozovski
,
G.
Itchaki
,
A.
Gafter-Gvili
,
C.
Leibovitch
,
P.
Raanani
,
H.
Ben-Zvi
,
M.
Szwarcwort
,
M.
Taylor-Abigadol
,
E. J.
Dann
, et al
2022
.
Humoral serological response to the BNT162b2 vaccine is abrogated in lymphoma patients within the first 12 months following treatment with anti-CD2O antibodies
.
Haematologica
107
:
715
720
.
26
Perry
,
C.
,
E.
Luttwak
,
R.
Balaban
,
G.
Shefer
,
M. M.
Morales
,
A.
Aharon
,
Y.
Tabib
,
Y. C.
Cohen
,
N.
Benyamini
,
O.
Beyar-Katz
, et al
2021
.
Efficacy of the BNT162b2 mRNA COVID-19 vaccine in patients with B-cell non-Hodgkin lymphoma
.
Blood Adv.
5
:
3053
3061
.
27
Shree
,
T.
,
V.
Shankar
,
J. J. K.
Lohmeyer
,
D. K.
Czerwinski
,
J. G.
Schroers-Martin
,
G. M.
Rodriguez
,
S.
Beygi
,
A. M.
Kanegai
,
K. S.
Corbelli
,
E.
Gabriel
, et al
2022
.
CD20-targeted therapy ablates de novo antibody response to vaccination but spares preestablished immunity
.
Blood Cancer Discov.
3
:
95
102
.
28
Eisenberg
,
R. A.
,
A. F.
Jawad
,
J.
Boyer
,
K.
Maurer
,
K.
McDonald
,
E. T.
Prak
,
K. E.
Sullivan
.
2013
.
Rituximab-treated patients have a poor response to influenza vaccination
.
J. Clin. Immunol.
33
:
388
396
.
29
Graalmann
,
T.
,
K.
Borst
,
H.
Manchanda
,
L.
Vaas
,
M.
Bruhn
,
L.
Graalmann
,
M.
Koster
,
M.
Verboom
,
M.
Hallensleben
,
C. A.
Guzmán
, et al
2021
.
B cell depletion impairs vaccination-induced CD8+ T cell responses in a type I interferon-dependent manner
.
Ann. Rheum. Dis.
80
:
1537
1544
.
30
R. B.
Muller
.,
R.
Maier
,
K.
Hoschler
,
M.
Zambon
,
B.
Ludewig
,
M.
Herrmann
,
H
,
Schulze-Koops
,
J.
von Kempis
.
2013
.
Efficient boosting of the antiviral T cell response in B cell–depleted patients with autoimmune rheumatic diseases following influenza vaccination
.
Clin. Exp. Rheumatol.
31
:
723
730
.
31
Nazi
,
I.
,
J. G.
Kelton
,
M.
Larché
,
D. P.
Snider
,
N. M.
Heddle
,
M. A.
Crowther
,
R. J.
Cook
,
A. T.
Tinmouth
,
J.
Mangel
,
D. M.
Arnold
.
2013
.
The effect of rituximab on vaccine responses in patients with immune thrombocytopenia
.
Blood
122
:
1946
1953
.
32
Squire
,
J. D.
,
A. Y.
Joshi
.
2022
.
Response to mRNA COVID-19 vaccination in three XLA patients
.
Vaccine
40
:
5299
5301
.
33
Apostolidis
,
S. A.
,
M.
Kakara
,
M. M.
Painter
,
R. R.
Goel
,
D.
Mathew
,
K.
Lenzi
,
A.
Rezk
,
K. R.
Patterson
,
D. A.
Espinoza
,
J. C.
Kadri
, et al
2021
.
Cellular and humoral immune responses following SARS-CoV-2 mRNA vaccination in patients with multiple sclerosis on anti-CD20 therapy
.
Nat. Med.
27
:
1990
2001
.
34
Riise
,
J.
,
S.
Meyer
,
I.
Blaas
,
A.
Chopra
,
T. T.
Tran
,
M.
Delic-Sarac
,
M. L.
Hestdalen
,
E.
Brodin
,
E. H.
Rustad
,
K. Z.
Dai
, et al
2022
.
Rituximab-treated patients with lymphoma develop strong CD8 T-cell responses following COVID-19 vaccination
.
Br. J. Haematol.
197
:
697
708
.
35
Stefanski
,
A. L.
,
H.
Rincon-Arevalo
,
E.
Schrezenmeier
,
K.
Karberg
,
F.
Szelinski
,
J.
Ritter
,
B.
Jahrsdorfer
,
H.
Schrezenmeier
,
C.
Ludwig
,
A.
Sattler
, et al
2022
.
B cell numbers predict humoral and cellular response upon SARS-CoV-2 vaccination among patients treated with rituximab
.
Arthritis Rheumatol.
74
:
934
947
.
36
Mrak
,
D.
,
S.
Tobudic
,
M.
Koblischke
,
M.
Graninger
,
H.
Radner
,
D.
Sieghart
,
P.
Hofer
,
T.
Perkmann
,
H.
Haslacher
,
R.
Thalhammer
, et al
2021
.
SARS-CoV-2 vaccination in rituximab-treated patients: B cells promote humoral immune responses in the presence of T-cell-mediated immunity
.
Ann. Rheum. Dis.
80
:
1345
1350
.
37
Alfonso-Dunn
,
R.
,
J.
Lin
,
V.
Kirschner
,
J.
Lei
,
G.
Feuer
,
M.
Malin
,
J.
Liu
,
M.
Roche
,
S. A.
Sadiq
.
2022
.
Strong T-cell activation in response to COVID-19 vaccination in multiple sclerosis patients receiving B-cell depleting therapies
.
Front. Immunol.
13
:
926318
.
38
Kedl
,
R. M.
,
E. W. Y.
Hsieh
,
T. E.
Morrison
,
G.
Samayoa-Reyes
,
S.
Flaherty
,
C. L.
Jackson
,
R.
Rochford
.
2023
.
Evidence for aerosol transfer of SARS-CoV-2-specific humoral immunity
.
Immunohorizons
7
:
307
309
.
39
Schultz
,
J. S.
,
M. K.
McCarthy
,
C.
Rester
,
K. R.
Sabourin
,
K.
Annen
,
M.
DomBourian
,
E.
Eisenmesser
,
A.
Frazer-Abel
,
V.
Knight
,
T.
Jaenisch
, et al
2021
.
Development and validation of a multiplex microsphere immunoassay using dried blood spots for SARS-CoV-2 seroprevalence: application in first responders in Colorado, USA
.
J. Clin. Microbiol.
59
:
e00290-21
.
40
Ritchie
,
M. E.
,
B.
Phipson
,
D.
Wu
,
Y.
Hu
,
C. W.
Law
,
W.
Shi
,
G. K.
Smyth
.
2015
.
limma powers differential expression analyses for RNA-sequencing and microarray studies
.
Nucleic Acids Res.
43
:
e47
.
41
Breslow
,
N.
1970
.
A generalized Kruskal–Wallis test for comparing k samples subject to unequal patterns of censorship
.
Biometrika
57
:
579
594
.
42
R: a language and environment for statistical Computing
.
Vienna, Austria
:
R Foundation for Statistical Computing
.
2022
. Available at: https://www.R-project.org/.
43
Grant
,
M. D.
,
K.
Bentley
,
C. A.
Fielding
,
K. M.
Hatfield
,
D. P.
Ings
,
D.
Harnum
,
E. C.
Wang
,
R. J.
Stanton
,
K. A.
Holder
.
2023
.
Combined anti-S1 and anti-S2 antibodies from hybrid immunity elicit potent cross-variant ADCC against SARS-CoV-2
.
JCI Insight
8
:
e170681
.
44
Rodda
,
L. B.
,
P. A.
Morawski
,
K. B.
Pruner
,
M. L.
Fahning
,
C. A.
Howard
,
N.
Franko
,
J.
Logue
,
J.
Eggenberger
,
C.
Stokes
,
I.
Golez
, et al
2022
.
Imprinted SARS-CoV-2-specific memory lymphocytes define hybrid immunity
.
Cell
185
:
1588
1601.e14
.
45
Fong
,
Y.
,
Y.
Huang
,
D.
Benkeser
,
L. N.
Carpp
,
G.
Áñ
ez
,
W.
Woo
,
A.
McGarry
,
L. M.
Dunkle
,
I.
Cho
,
C. R.
Houchens
, et al
Immune Assays Team; Coronavirus Vaccine Prevention Network/2019nCoV-301 Principal Investigators and Study Team; U.S. Government/CoVPN Biostatistics Team
.
2023
.
Immune correlates analysis of the PREVENT-19 COVID-19 vaccine efficacy clinical trial
.
Nat. Commun.
14
:
1581
.
46
Gilbert
,
P. B.
,
R. O.
Donis
,
R. A.
Koup
,
Y.
Fong
,
S. A.
Plotkin
,
D.
Follmann
.
2022
.
A COVID-19 milestone attained—a correlate of protection for vaccines
.
N. Engl. J. Med.
387
:
2203
2206
.
47
Ferguson
,
J.
,
K.
Murugesan
,
N.
Banaei
,
A.
Liu
.
2021
.
Interferon-gamma release assay testing to assess COVID-19 vaccination response in a SARS-CoV-2 seronegative patient on rituximab: a case report
.
Int. J. Infect. Dis.
110
:
229
231
.
48
Gadani
,
S. P.
,
M.
Reyes-Mantilla
,
L.
Jank
,
S.
Harris
,
M.
Douglas
,
M. D.
Smith
,
P. A.
Calabresi
,
E. M.
Mowry
,
K. C.
Fitzgerald
,
P.
Bhargava
.
2021
.
Discordant humoral and T cell immune responses to SARS-CoV-2 vaccination in people with multiple sclerosis on anti-CD20 therapy
.
EBioMedicine
73:
103636
.
49
Madelon
,
N.
,
K.
Lauper
,
G.
Breville
,
I.
Sabater Royo
,
R.
Goldstein
,
D. O.
Andrey
,
A.
Grifoni
,
A.
Sette
,
L.
Kaiser
,
C. A.
Siegrist
, et al
2022
.
Robust T-cell responses in anti-CD20-treated patients following COVID-19 vaccination: a prospective cohort study
.
Clin. Infect. Dis.
75
:
e1037
e1045
.
50
Moor
,
M. B.
,
F.
Suter-Riniker
,
M. P.
Horn
,
D.
Aeberli
,
J.
Amsler
,
B.
Möller
,
L. M.
Njue
,
C.
Medri
,
A.
Angelillo-Scherrer
,
L.
Borradori
, et al
2021
.
Humoral and cellular responses to mRNA vaccines against SARS-CoV-2 in patients with a history of CD20 B-cell-depleting therapy (RituxiVac): an investigator-initiated, single-centre, open-label study
.
Lancet Rheumatol.
3
:
e789
97
.
51
Tallantyre
,
E. C.
,
N.
Vickaryous
,
V.
Anderson
,
A. N.
Asardag
,
D.
Baker
,
J.
Bestwick
,
K.
Bramhall
,
R.
Chance
,
N.
Evangelou
,
K.
George
, et al
2022
.
COVID-19 vaccine response in people with multiple sclerosis
.
Ann. Neurol.
91
:
89
100
.
52
Mei
,
H. E.
,
D.
Frölich
,
C.
Giesecke
,
C.
Loddenkemper
,
K.
Reiter
,
S.
Schmidt
,
E.
Feist
,
C.
Daridon
,
H.-P.
Tony
,
A.
Radbruch
,
T.
Dörner
.
2010
.
Steady-state generation of mucosal IgA+ plasmablasts is not abrogated by B-cell depletion therapy with rituximab
.
Blood
116
:
5181
5190
.
53
He
,
Y.
,
M.
Shimoda
,
Y.
Ono
,
I. B.
Villalobos
,
A.
Mitra
,
T.
Konia
,
S. A.
Grando
,
J. J.
Zone
,
E.
Maverakis
.
2015
.
Persistence of autoreactive IgA-secreting B cells despite multiple immunosuppressive medications including rituximab
.
JAMA Dermatol.
151
:
646
650
.
54
Lucchini
,
M.
,
A.
Bianco
,
P.
Del Giacomo
,
C.
De Fino
,
V.
Nociti
,
M.
Mirabella
.
2020
.
Is serological response to SARS-CoV-2 preserved in MS patients on ocrelizumab treatment? A case report
.
Mult. Scler. Relat. Disord.
44
:
102323
.
55
McMahan
,
K.
,
J.
Yu
,
N. B.
Mercado
,
C.
Loos
,
L. H.
Tostanoski
,
A.
Chandrashekar
,
J.
Liu
,
L.
Peter
,
C.
Atyeo
,
A.
Zhu
, et al
2021
.
Correlates of protection against SARS-CoV-2 in rhesus macaques
.
Nature
590
:
630
634
.
56
McMahan
,
K.
,
F.
Wegmann
,
M.
Aid
,
M.
Sciacca
,
J.
Liu
,
N. P.
Hachmann
,
J.
Miller
,
C.
Jacob-Dolan
,
O.
Powers
,
D.
Hope
, et al
2024
.
Mucosal boosting enhances vaccine protection against SARS-CoV-2 in macaques
.
Nature
626
:
385
391
.
57
Isho
,
B.
,
K. T.
Abe
,
M.
Zuo
,
A. J.
Jamal
,
B.
Rathod
,
J. H.
Wang
,
Z.
Li
,
G.
Chao
,
O. L.
Rojas
,
Y. M.
Bang
, et al
2020
.
Persistence of serum and saliva antibody responses to SARS-CoV-2 spike antigens in COVID-19 patients
.
Sci. Immunol.
5
:
eabe5511
.
58
Sheikh-Mohamed
,
S.
,
E. C.
Sanders
,
J. L.
Gommerman
,
M. C.
Tal
.
2022
.
Guardians of the oral and nasopharyngeal galaxy: IgA and protection against SARS-CoV-2 infection
.
Immunol. Rev.
309
:
75
85
.
59
Barouch
,
D. H.
2022
.
COVID-19 vaccines—immunity, variants, boosters
.
N. Engl. J. Med.
387
:
1011
1020
.
60
Liu
,
J.
,
A.
Chandrashekar
,
D.
Sellers
,
J.
Barrett
,
C.
Jacob-Dolan
,
M.
Lifton
,
K.
McMahan
,
M.
Sciacca
,
H.
VanWyk
,
C.
Wu
, et al
2022
.
Vaccines elicit highly conserved cellular immunity to SARS-CoV-2 Omicron
.
Nature
603
:
493
496
.
61
Dan
,
J. M.
,
J.
Mateus
,
Y.
Kato
,
K. M.
Hastie
,
E. D.
Yu
,
C. E.
Faliti
,
A.
Grifoni
,
S. I.
Ramirez
,
S.
Haupt
,
A.
Frazier
, et al
2021
.
Immunological memory to SARS-CoV-2 assessed for up to 8 months after infection
.
Science
371
:
eabf4063
.
62
Grifoni
,
A.
,
D.
Weiskopf
,
S. I.
Ramirez
,
J.
Mateus
,
J. M.
Dan
,
C. R.
Moderbacher
,
S. A.
Rawlings
,
A.
Sutherland
,
L.
Premkumar
,
R. S.
Jadi
, et al
2020
.
Targets of T cell responses to SARS-CoV-2 coronavirus in humans with COVID-19 disease and unexposed individuals
.
Cell
181
:
1489
1501.e15
.
63
Murray
,
S. M.
,
A. M.
Ansari
,
J.
Frater
,
P.
Klenerman
,
S.
Dunachie
,
E.
Barnes
,
A.
Ogbe
.
2023
.
The impact of pre-existing cross-reactive immunity on SARS-CoV-2 infection and vaccine responses
.
Nat. Rev. Immunol.
23
:
304
316
.
64
Routhu
,
N. K.
,
S. D.
Stampfer
,
L.
Lai
,
A.
Akhtar
,
X.
Tong
,
D.
Yuan
,
T. M.
Chicz
,
R. P.
McNamara
,
K.
Jakkala
,
M. E.
Davis-Gardner
, et al
2023
.
Efficacy of mRNA-1273 and Novavax ancestral or BA.1 spike booster vaccines against SARS-CoV-2 BA.5 infection in nonhuman primates
.
Sci. Immunol.
8
:
eadg7015
.
This article is distributed under the terms of the CC BY-NC 4.0 Unported license.

Supplementary data