Abstract
Background:
People with MS (PwMS) and related conditions treated with anti-CD20 and S1P modulating therapies exhibit attenuated immune responses to SARS-CoV-2 vaccines. It remains unclear whether humoral/T-cell responses are valid surrogates for postvaccine immunity.
Objective:
To characterize COVID-19 vaccine-breakthrough infections in this population.
Methods:
We conducted a prospective multicenter cohort study of PwMS and related CNS autoimmune conditions with confirmed breakthrough infections. Postvaccination antibody response, disease-modifying therapies (DMTs) at the time of vaccination, and DMT at the time of infection were assessed.
Results:
Two hundred nine patients had 211 breakthrough infections. Use of anti-CD20 agents at time of infection was associated with increased infection severity (p = 0.0474, odds ratio (OR) = 5.923) for infections during the Omicron surge and demonstrated a trend among the total cohort (p = 0.0533). However, neither use of anti-CD20 agents at the time of vaccination nor postvaccination antibody response was associated with hospitalization risk. Anti-CD20 therapies were relatively overrepresented compared to a similar prevaccination-era COVID-19 cohort.
Conclusion:
Use of anti-CD20 therapies during vaccine breakthrough COVID-19 infection is associated with higher severity. However, the attenuated postvaccination humoral response associated with anti-CD20 therapy use during vaccination may not drive increased infection severity. Further studies are necessary to determine if this attenuated vaccine response may be associated with an increased likelihood of breakthrough infection.
Keywords: Disease-modifying therapies, multiple sclerosis, immunology
Introduction
People with multiple sclerosis (MS) (PwMS) and related conditions use disease-modifying therapies (DMTs) with immunomodulatory or immunosuppressive effects. At the onset of the COVID-19 pandemic, the MS community was concerned that immunosuppressive medications, including cell-depleting and -sequestering therapies, might substantially elevate the risk of severe infections. Multiple early, prevaccination-era studies demonstrated that, although patients treated with anti-CD20 therapies such as ocrelizumab or rituximab may have higher rates of hospitalization, most PwMS and related conditions have favorable COVID-19 outcomes despite immunosuppression.1–5 The original New York COVID-19 Neuroimmunology Consortium (NYNIC) collaboration showed 1 that the majority of MS patients on anti-CD20 agents exhibited mild-to-moderate COVID-19 course. However, unlike the vast majority of those using noncell-depleting therapies who produced measurable postinfection antibodies, only a minority of those using cell-depleting agents produced antibodies postinfection. This attenuation in humoral responses to COVID-19 infection has been demonstrated in other cohorts.6–9 A similarly poor antibody response has been found post-SARS-CoV-2-vaccination in patients treated with anti-CD20 therapies and sphingosine-1-phosphate (S1P) receptor modulators,10–16 a trend seen previously with responses to other vaccines.17–20 Postvaccination SARS-CoV-2 humoral responses are generally lower in magnitude and delayed in PwMS treated with these agents, and this attenuation persists after multiple vaccine doses. Reported SARS-CoV-2 spike-protein-specific seropositivity rates with anti-CD20 therapies following a primary course of two mRNA vaccine doses vary widely across studies, ranging from 13% to 89%, rising only slightly following a third vaccine dose.11,14,15,21–27 Similarly, reported seropositivity after vaccination in PwMS treated with S1P agents ranged widely from 3.8% to 85.7%, with only a small rise after a third dose.10,14,15,25–27 Variations in reported seropositivity rates may be due to differences in assays used, degree of cell depletion, and/or timing of infused medications and vaccination. While SARS-CoV-2 spike protein-specific T-cell responses are preserved in PwMS treated with anti-CD20 therapies,11,21,22,25–28 raising the possibility of at least partially intact vaccine-mediated protection, the degree and spectrum of T-cell responses (CD4 and CD8) differed in patients with and without neutralizing antibodies. 11 In the setting of DMT use, it remains unclear to what extent either humoral or T-cell responses are valid surrogates for effective postvaccine immunity, from the standpoints of both incidence and severity of breakthrough infections. Clinical data establishing the incidence and severity of COVID-19 infection in vaccinated PwMS and related conditions is needed to pragmatically assess vaccine efficacy in this population.
The NYCNIC registry aimed at analyzing hospitalization risk in a cohort of PwMS and related conditions (NYCNIC-2) who experienced vaccine-breakthrough infections.
Methods
Data collection
We conducted a multicenter, observational cohort study of adults with MS and related CNS autoimmune inflammatory disorders (neuromyelitis spectrum disorder (NMOSD), myelin oligodendrocyte glycoprotein-associated disease (MOGAD), neurosarcoidosis, or autoimmune encephalomyelitis), who had confirmed COVID-19 diagnoses (using either PCR or antigen nasal swab) occurring at least 2 weeks after full vaccination (i.e. “vaccine-breakthrough infections”). Full vaccination was defined as at least two doses of the Pfizer-BioNTech or Moderna mRNA vaccines or at least one dose of the Janssen vaccine.
Five MS centers in New York City and the surrounding vicinity participated: the MS centers at Columbia University Irving Medical Center, Icahn School of Medicine at Mount Sinai, New York University Langone Medical Center, Neurological Associates of Long Island, and Northwell Health Partners. All treating neurologists at participating centers were instructed to inquire and record breakthrough COVID infections during clinic visits or during a phone call report from patients (this was a routine practice at all sites). Treating neurologists recorded cases in real-time and were responsible for subsequently following up on all COVID outcomes until infection resolution, recording the final outcome. The authors (designated neurologists at each site) were responsible for data entry and completion.
The study was approved by Institutional Review Boards at each participating institution. Data were collected from medical records of patients between 1 February 2020 and 31 January 2022.
Procedures
Due to substantial differences in breakthrough rates of SARS-CoV-2 strains, infections were stratified as occurring “pre-Omicron” if prior to 1 December 2021, or “During Omicron” if occurring between 1 December 2021 and 31 January 2022.
Outcome definitions
The primary outcome of this study was hospitalization status of patients with COVID-19 infection. The severity of infection was measured on a 4-point ordinal scale: 1. home care; 2. hospitalization without mechanical ventilation; 3. hospitalization and mechanical ventilation, and 4. death. Potential risk factors such as patients’ demographics (age and sex) and medical history/clinical characteristics (time interval from vaccination or prior infection to vaccine-breakthrough infection, receipt of booster or additional doses of vaccines, and cardiovascular comorbidities) were recorded. Hypertension, hyperlipidemia, diabetes, coronary artery disease or stroke, BMI ⩾ 40, and active tobacco use were classified as cardiovascular risk factors. Subjects were grouped into either having no cardiovascular risk factors or having at least one cardiovascular risk factor. Neuroimmune disease-specific data included: duration of disease; DMT use at the time of vaccination; DMT use at the time of vaccine-breakthrough infection; and baseline ambulatory status (measured along a 5-point ordinal scale: 1. fully ambulatory, 2. limited ambulation with no assistance, 3. unilateral assistance, 4. bilateral assistance and 5. wheelchair/bedbound) were collected. DMTs were grouped by mechanism of action (see Table 1). If the vaccine-breakthrough infection occurred after a prior confirmed SARS-CoV-2 infection, it was labeled as a “re-infection.” Use of treatments for COVID-19 infection was collected, as these could impact infection severity and could potentially confound the relationship between DMT type and infection outcome.
Table 1.
Baseline demographic and clinical characteristics (registry of breakthrough infections).
| Breakthrough infections | Pre-Omicron, N = 45 | During Omicron, N = 166 | Total, N = 211 | |
|---|---|---|---|---|
| Sex | Female | 32 (71%) | 117 (70%) | 149 (71%) |
| Male | 13 (29%) | 49 (30%) | 62 (29%) | |
| Age (years) | Median | 41.5 | 42 | 42 |
| IQR | 19 | 26 | 20 | |
| Race | White/Caucasian | 33 (73%) | 105 (63%) | 138 (65%) |
| Black/African American | 4 (9%) | 31 (19%) | 35 (17%) | |
| Asian | 0 (0%) | 2 (1%) | 2 (1%) | |
| Other | 6 (13%) | 12 (7%) | 18 (9%) | |
| Unknown | 2 (4%) | 16 (10%) | 18 (9%) | |
| Ethnicity | Hispanic/Latino | 8 (18%) | 37 (22%) | 45 (21%) |
| Non-Hispanic/Latino | 32 (71%) | 115 (69%) | 147 (70%) | |
| Unknown | 5 (11%) | 14 (8%) | 19 (9%) | |
| Disease type | MS (RIS, CIS, RRMS, SPMS, PPMS) | 194 (91.9%) | ||
| NMOSD | 11 (5.2%) | |||
| Other (TM, MOGAD, Other, Hypophysitis, anti-GAD syndrome) | 6 (2.9%) | |||
| Gait assistance | None | 37 (82%) | 149 (90%) | 186 (88%) |
| Unilateral/Bilateral Non-ambulatory | 8 (18%) | 17 (10%) | 25 (12%) | |
| Vascular comorbidities | No | 30 (66%) | 122 (73%) | 152 (72%) |
| Yes | 15 (34%) | 44 (27%) | 59 (28%) | |
| Time between last vaccine dose and COVID-19 infection (months) | Median (min–max) |
4.75 (0.25–11.0) |
4.00 (0.25–12.00) |
4.25 (0.25–12.0) |
| COVID treatment | No | 24 (53%) | 127 (77%) | 151 (72%) |
| Yes | 21 (47%) | 39 (23%) | 60 (28%) | |
| Infection severity |
Asymptomatic | 1 (2.2%) | 3 (1.8%) | 4 (1.9%) |
| Home care | 37 (82.2%) | 148 (89.2%) | 185 (87.7%) | |
| Hospitalized non-ICU | 6 (13.3%) | 13 (7.8%) | 19 (9%) | |
| Critically ill—survived | 0 (0%) | 1 (0.6%) | 1 (0.5%) | |
| Death | 1 (2.2%) | 1 (0.6%) | 2 (1%) | |
| DMT class during vaccination | Anti-CD20 | 30 (67%) | 93 (56%) | 123 (58%) |
| S1P modulator | 4 (9%) | 15 (9%) | 19 (9%) | |
| Fumarate | 3 (7%) | 14 (8%) | 17 (8%) | |
| NTZ | 1 (2%) | 15 (9%) | 16 (8%) | |
| TFL | 1 (2%) | 2 (1%) | 3 (1%) | |
| Platform (IFN, GA) | 1 (2%) | 3 (2%) | 4 2%) | |
| None | 4 (9%) | 21 (13%) | 25 (12%) | |
| Other | 1 (2%) | 3 (2%) | 4 (2%) | |
| DMT class during infection | Anti-CD20 | 30 (67%) | 103 (62%) | 133 (63%) |
| S1P modulator | 4 (9%) | 13 (8%) | 17 (8%) | |
| Fumarate | 3 (7%) | 14 (8%) | 17 (8%) | |
| NTZ | 3 (7%) | 15 (9%) | 18 (9%) | |
| TFL | 1 (2%) | 2 (1%) | 3 (1%) | |
| Platform (IFN, GA) | 1 (2%) | 2 (1%) | 3 (1%) | |
| None | 2 (4%) | 11 (7%) | 13 (6%) | |
| Other | 1 (2%) | 6 (4%) | 7 (3%) | |
CIS: clinically isolated syndrome; DMT: disease-modifying therapy; MOGAD: myelin oligodendrocyte glycoprotein associated disease; MS: multiple sclerosis; NMOSD: neuromyelitis optica spectrum disorder; PPMS: primary progressive MS; RIS: radiologically isolated syndrome; SPMS: secondary progressive MS; TM: transverse myelitis.
Platform agents: Glatiramer acetate, Interferons.
Other therapies: Mycophenolate mofetil (2), Azathioprine (1), Intravenous immunoglobulins (2), Satralizumab (1).
Postvaccination antibody data were obtained from readily available clinical (standard-of-care) antibody testing for a subset of patients. Three assays were used across the participating sites: Roche Elecsys® Anti-SARS-CoV-2 S (RBD) Assay, 29 SeroKlir Kantaro Semi-Quantitative SARS-CoV-2 IgG Antibody Kit, or Diasorin SARS-CoV-2 TrimericS IgG assay. Quantitative antibody data were compiled into binary “Within Normal” or “Abnormally Low” values. For the Roche Elecsys Assay, a value of >50 U/mL was labeled as “Within Normal,” while ⩽50 U/mL (⩽20% upper limit of detection) was labeled as “Abnormally Low,” consistent with a precedent set by recent National Institutes of Health clinical trials evaluating vaccine nonresponders.30,31 Value of >25 AU/mL on the SeroKlir assay was labeled as “Within Normal,” while ⩽25 AU/mL (⩽20% upper limit of detection) was considered “Abnormally Low.” On the Diasorin assay, ⩾33.8 was labeled “Within Normal” and <33.8 was labeled “Abnormally Low.”
Statistical analyses
Descriptive statistics (frequency distribution for categorical variables and mean, SD, median, interquartile range, minimum, and maximum for continuous variables) were calculated.
Univariable logistic regression was used to screen variables with a p-value criterion of p < 0.05 for entry into the model selection procedure. Stepwise selection was used with variable entry and retention criteria of p < 0.05 to select the final multivariable model. Variables that were specified for clinical relevance were retained in the model at each selection step. Fit statistics (Akaike Information Criterion and Schwarz Criterion) were also used in the model selection procedure. Firth’s correction was used in cases where quasi-separation was present in the model due to categorical levels with low frequencies.
All statistics were performed using SAS (Statistical Analysis System) Version 9.4.
Results
Demographics
See Table 1 for patient demographics and clinical characteristics. Our cohort included 209 patients with 211 vaccine-breakthrough infections, of which 17 were re-infections. Forty-five vaccine-breakthrough infections occurred pre-Omicron (i.e. prior to 12 January 2021) and 166 occurred during the initial Omicron (BA.1) surge (between 12 January 2021 and 31 January 2022). Out of vaccine-breakthrough infections that were also re-infections, one occurred Pre-Omicron and 16 occurred during Omicron.
In the overall cohort of breakthrough infections, the median age was 42 (range, 19–78). Seventy-one percent of infections occurred in females, 65% in Caucasians, 17% in black or African Americans, and 21% among Hispanics or Latinos. The median disease duration was 9 years (IQR: 5–16 years). Most infections (91.9%) occurred in the setting of MS phenotype diagnoses (clinically definite MS, clinically isolated syndrome, or radiologically isolated syndrome). The majority (88%) occurred in fully ambulatory patients, and 72% in patients with no cardiovascular comorbidity.
One hundred nineteen subjects received Pfizer, 51 received Moderna, 9 received Janssen, and 30 were vaccinated with an unknown vaccine type. Seventy-three subjects were confirmed to have received a “booster” prior to infection (i.e. a third dose following initial mRNA vaccines or a second vaccination following an initial Janssen administration). Boosters had been received by 45 of 133 (34%) subjects treated with anti-CD20 agents, 9 of 17 (53%) subjects treated with S1P modulators, and 18 of 61 (30%) subjects treated with other or no treatments at the time of documented new infection. None of the patients included in the study had received EvusheldTM (tixagevimab and cilgavimab). 32
In both the pre-Omicron and during Omicron groups, the median times for breakthrough infections were 4 to 5 months after the last vaccine dose. The median time to vaccine-breakthrough infections that were also re-infections was 4.5 months after the last vaccine dose and 11 months following the prior infection.
DMT profiles at time of vaccination and infection
One hundred ninety-eight vaccine breakthrough infections occurred during treatment with a DMT. Out of 211 infections, 190 occurred during treatment with the same DMT class that was used at the time of vaccination.
Among the 17 vaccine-breakthrough that were also re-infections, 16 occurred during treatment with the same DMT class used at the time of vaccination and prior infection.
SARS-CoV-2 vaccine-breakthrough infection outcomes
Of the 211 breakthrough infections, 4 (1.9%) were asymptomatic, 22 (10.4%) required hospitalization (7 (15.5%) in the pre-Omicron group, and 15 (9.0%) in the Omicron group), 3 (1.4%) infections caused critical illness (ICU care or outcome of death), and 2 (1%) were fatal (Figure 1).
Figure 1.
Clinical outcomes of vaccine-breakthrough COVID-19 infections. A large majority of infections were mild and did not require hospitalization in both the “Pre-Omicron” (a) and the “During Omicron” (b) time periods. The rate of hospitalization “During-Omicron” (9.0%) was numerically lower than the rate of hospitalization “Pre-Omicron” (15.5%), consistent with a generally milder illness. Rates of critical illness and death were very low in both groups. One patient was critically ill and died in the “Pre-Omicron” group. Two patients were critically ill, one of whom died in the “During Omicron” group.
DMT use and severity of SARS-CoV-2 vaccine-breakthrough infection
Of the 22 infections that led to hospitalization, 17 (77%) occurred in the setting of anti-CD20 treatment, 2 (9%) occurred in the setting of S1P modulator treatment, 1 (5%) during teriflunomide treatment, and 2 (9%) in untreated patients. The 3 infections that led to critical illness or death all occurred in the setting of anti-CD20 treatment.
To assess whether DMT at the time of infection is associated with COVID-19 severity, variables including age, sex, ambulatory function, comorbidities, vaccine type, booster status, time since last vaccination, and DMT class at the time of infection were analyzed for association with hospitalization for COVID-19 infection. Univariable analysis showed that age and cardiovascular comorbidities were associated with hospitalization. Model selection retained age in the final model. Multivariable analysis showed that controlling for age, anti-CD20 use at the time of infection predicted hospitalization during the Omicron era (p = 0.0474, OR 5.923), and a model demonstrated a similar trend but failed to meet significance when analyzing the total cohort (p = 0.0553). The odds of hospitalization for COVID-19 infection for patients being treated with anti-CD20 agents during the Omicron surge was 5.923 times higher (95% CI: 1.021–34.37) than those on other or no therapies, excluding S1P modulators (Table 2).
Table 2.
DMT type as predictor of hospitalization for SARS-CoV-2 infection.
| Pre-Omicron (95% CI) |
During Omicron (95% CI) |
Total (95% CI) |
|
|---|---|---|---|
| DMT type during infection | |||
| Anti-CD20 therapies versus other DMT (except S1P modulators) |
p = 0.9067 OR 0.886 (0.117–6.723) |
p = 0.0474, OR 5.923 (1.021–34.373) |
p = 0.0553 OR 3.394 (0.973–11.845) |
| DMT type during vaccination | |||
| Anti-CD20 therapies versus other DMT (except S1P modulators) |
p = 0.9067 OR 0.886 (0.117–6.723) |
p = 0.0842 OR 3.467 (0.846–14.213) |
p = 0.0987 OR 2.544 (0.840–7.709) |
CI: confidence interval; DMT: disease-modifying therapy; OR: odds ratio.
“Other DMT”: Fumarates, Natalizumab, Teriflunomide, Interferons, Glatiramer Acetate, Azathioprine.Significant p value bold.
Similarly, to determine whether attenuated postvaccine immune response plays a role in the risk of hospitalization, a multivariable analysis was performed to evaluate a potential relationship between infection severity and DMT class at the time of vaccination. Anti-CD20 therapy use at the time of vaccination was not associated with hospitalization for breakthrough infection (pre-Omicron p = 0.9067, Omicron p = 0.0842, and total p = 0.0987).
There were no associations between risk of hospitalization with any other DMT, including S1P modulators.
Treatments for SARS-CoV-2 infection across DMTs
Although detailed evaluation of the use of specific COVID-19 treatments was out of the scope of the study, a general analysis of COVID-19 treatment utilization (yes or no) across DMT classes was included to determine whether such treatments were used more frequently by patients undergoing therapy with certain DMTs. Treatments for COVID-19 infection in our cohort consisted of varied regimens that included systemic corticosteroids, antivirals, antibiotics, and monoclonal antibodies. Fifty-six infections were treated with at least one of the above agents and 151 infections were not treated (4 unknown). Forty-one out of 133 (31%) infections occurring during anti-CD20 therapy were treated, 0 out of 17 (0%) infections during S1P modulating therapy were treated, and 11 out of 61 (18%) infections occurring in the setting of other or no DMT were treated. The same trend remained when limiting the analysis to nonhospitalized infections only, demonstrating that increased hospitalization of patients on anti-CD20 therapies was not the reason behind the increased utilization of COVID-19 treatments.
Pre-infection antibody status and severity of SARS-CoV-2 vaccine-breakthrough infection
We next performed an exploratory analysis to determine whether pre-infection standard-of-care SARS-CoV-2 antibody status predicts vaccine-breakthrough infection severity.
Seventy-nine patients had known antibody status prior to vaccine-breakthrough infection, with more than 80% having a known value within 6 months of breakthrough infection. Median time from antibody testing to new infection was 3 months, a time frame during which antibody levels are relatively stable.33,34
“Abnormally Low” values were exhibited by 35 out of 46 (76%) patients treated with anti-CD20 therapies at the time of vaccination, 2 out of 6 (33%) treated with S1P modulators at the time of vaccination, and 1 out of 28 (3.6%) of all others (Figure 2). Among the 79 patients with available antibody data, there were 8 hospitalizations, all in the setting of anti-CD20 therapy, six of which were in the setting of “Abnormally Low” antibody levels, and one of which led to death.
Figure 2.
Preinfection SARS-CoV-2 antibody status prior to vaccine-breakthrough infection. Seventy-nine patients had known antibody levels prior to vaccine-breakthrough COVID-19 infection, as assessed by one of three different assays. Given the variety of assays, quantitative antibody data was assigned a binary value of either “Abnormally Low” or “Within Normal” to simplify the analysis (see section on antibody status in “Results”). “Abnormally Low” values were exhibited by 76% of patients who had been treated with anti-CD20 therapies at the time of vaccination, 33% of those who had been treated with S1P modulators at the time of vaccination, and 3.6% of all others in the cohort.
In a multivariable model accounting for age, sex, ambulatory function, comorbidities, vaccine type, booster status, and time since the last vaccination, there was no difference in risk of hospitalization for COVID-19 infection between patients with normal antibody levels and those with abnormally low antibody levels in the total cohort (p = 0.1270). Subset analysis isolated to patients on either anti-CD20 or S1P modulating therapies (p = 0.5004), and those on anti-CD20 therapies alone (p = 0.6446) similarly did not show a relationship between the antibody level and severity of infection (Table 3).
Table 3.
Abnormally low antibody level as predictor of hospitalization for SARS-CoV-2 infection.
| DMT type during vaccination | p value | OR (95% CI) |
|---|---|---|
| All DMTs | 0.1270 | 3.343 (0.710–15.752) |
| Anti-CD20 therapies and S1P modulators | 0.5004 | 1.987 (0.270–14.633) |
| Anti-CD20 therapies alone | 0.6446 | 1.650 (0.196–13.874) |
CI: confidence interval; DMT: disease-modifying therapy; OR: odds ratio.
“All DMTs”: Anti-CD20 therapies, S1P Modulators, Fumarates, Natalizumab, Teriflunomide, Interferons, Glatiramer Acetate, Azathioprine, IVIG, Mycophenolate Mofetil, Satralizumab, or No DMT.
DMT representation in vaccine-breakthrough cohort
There was a relative overrepresentation of anti-CD20 therapy use in our cohort, which was higher than what we expected based on internal estimates of anti-CD20 therapy utilization. 1 One hundred thirty-three (63%) of the total 211 vaccine-breakthrough infections, 30 (67%) of the 45 pre-Omicron infections, and 103 (62%) of 166 “during-Omicron” infections occurred during anti-CD20 therapy. One hundred twenty-three (58%) of the total vaccine-breakthrough infections, 30 (67%) of the 45 pre-Omicron infections, and 93 (56%) of the 166 “during-Omicron” infections occurred in patients who had been vaccinated while on anti-CD20 therapy. Finally, of the 17 vaccine-breakthrough infections that were also re-infections, 12 (71%) occurred in patients who had been previously infected and vaccinated while being treated with anti-CD20 agents.
To further study the occurrence of COVID-19 infections in unvaccinated and vaccinated patients on anti-CD20 agents, we utilized our prior observational data from the NYCNIC1 cohort. 1 Among unvaccinated patients in the NYCNIC1 cohort, 35% reported infections occurred in patients treated with anti-CD20 therapies. 1 In comparison, 63% of vaccine-breakthrough infections in the present NYCNIC2 cohort occurred in patients treated with anti-CD20 therapies (p < 0.0001). When limiting the analysis to only subjects with confirmed infection (PCR or antibody-positive only) from NYCNIC1, infections during anti-CD20 therapy use account for only 30.1% of the total, and similar statistically significant differences are observed. This apparent relative over-representation was not present with other DMT categories (Figure 3).
Figure 3.
DMT representation in the vaccine-breakthrough cohort. There was a relative over-representation of anti-CD20 therapy use in our cohort, with 63% of infections in the cohort occurring in the setting of anti-CD20 therapy use, 8% occurring in the setting of S1P modulating therapy, and 29% in the setting of other DMTs or no DMT. We compared these values to those of the pre-vaccine era NYCNIC1 cohort. The representation of anti-CD20 agents in the vaccine-breakthrough NYCNIC2 cohort (63%) was higher than in the pre-vaccine NYCNIC1 cohort (35%). Consequently, the representation of other DMTs in the NYCNIC2 cohort was smaller than in NYCNIC1. The representation of S1P modulating therapies in both cohorts did not differ. ****p < 0.0001.
Discussion
To our knowledge, this is the first large multicenter study to examine COVID-19 clinical outcomes in PwMS and related conditions in the postvaccination era. While other studies have evaluated how DMTs may affect biomarkers of vaccine response (postvaccination antibody production and/or T-cell function),2,10–16,22,24,28,35 we assessed the real-world clinical impact of vaccination in a large cohort of PwMS and related conditions, before and during the first Omicron surge.
We assessed whether the risk of hospitalization for COVID-19 vaccine-breakthrough infection is related to DMT class at the time of infection or at the time of vaccination. Our results showed that the use of anti-CD20 agents at the time of infection is associated with increased severity, as measured by the risk of hospitalization, and this risk is further increased with age. While similar findings were demonstrated in early pre-vaccine-era studies,1,6–9 ours is the first to describe this in vaccinated individuals. As hospitalization risk increased with both anti-CD20 therapy use and older age, elderly patients treated with such agents should be counseled on their risk and proper precautions. Despite the increased risk of hospitalization with anti-CD20 use at the time of infection, it is reassuring that the great majority of anti-CD20-treated infections did not require hospitalization. There was higher COVID-19 treatment utilization for infections treated in the setting of anti-CD20 therapy, which may have somewhat attenuated the observed increased risk of hospitalization. This higher utilization may be due to prescribing practices arising at least in part from reported increased pre-vaccine COVID-19 severity with anti-CD20 therapies.2–5
We next assessed whether reduced postvaccine immune response increases the risk of hospitalization. While most patients who were treated with anti-CD20 agents at the time of vaccination did not mount adequate antibody responses, use of anti-CD20 agents at the time of vaccination was not associated with increased hospitalization risk, suggesting that attenuated antibody production in this population may not lead to increased infection severity. Consistent with this notion, we found no association between inadequate postvaccine antibody response and infection severity in exploratory analysis of a subgroup with known antibody levels tested as part of standard-of-care. It is possible that intact T-cell function in this population may offer some vaccine-related protection against severe infection.11,12,14,21,22
While this study was designed to assess the severity, and not incidence, of infection, there was an overrepresentation of anti-CD20 therapy use by patients with infections in our cohort, nearly double the proportion found in our pre-vaccine NYCNIC1 cohort. 1 While we cannot rule out that this finding is related to changes in prescribing practices, patient habits, or reporting bias, these factors alone seem unlikely to account for the magnitude of overrepresentation. Furthermore, this over-abundance was numerically higher in the pre-Omicron era, in which breakthrough infections were rarer in the setting of normal immune function, and it was even higher among vaccine-breakthrough infections that were also re-infections. Our results suggest the possibility that reduced vaccine response may drive increased breakthrough infection frequency in patients vaccinated in the setting of anti-CD20 therapy. Larger-scale population studies are necessary to further assess this possibility.
Our study had several limitations. While the total cohort was large, the relative over-representation of anti-CD20 therapies may have limited the power to demonstrate significant associations with other DMTs. The relatively small number of hospitalizations may have led to limited abilities to detect differences in outcomes. Antibody levels were obtained from clinically available standard-of-care testing data and available only in a subset of patients, which limited the power of our exploratory analysis. While the timing interval between antibody testing and breakthrough infection was variable, it was less than 6 months in 80% of the subset, a time frame during which antibody levels are generally stable.33,34 As with any registry study, there is a risk of reporting bias and underreporting of mild COVID-19 infections. However, as data were collected prospectively by neurologists at only five centers and a designated neurologist at each site was tasked with ensuring record completion, our study likely has higher data quality than other larger registry studies using less rigorous methodology. Finally, though it is conceivable that precautions such as mask wearing and social distancing could have influenced the incidence of infection in our cohort, analysis of such factors was outside the scope of our study. However, we posit that patients treated with anti-CD20 agents would have been more likely to take such precautions, as many of them knew that anti-CD20 agents were associated with both an increased infection severity in pre-vaccine cohorts as well as impaired vaccine-mediated antibody production. Hence, bias related to such factors might lead to decreased representation of anti-CD20 agents in our cohort, not overrepresentation.
Our study demonstrates that the increased risk of infection severity associated with anti-CD20 agents in prevaccine studies persists in the postvaccine era. Neither the type of DMT at vaccination nor antibody levels were associated with infection severity. The overrepresentation of anti-CD20 use in our cohort suggests that postvaccine antibody response could be related to infection incidence. Further studies are necessary to delineate this possible relationship, especially in the setting of novel variants and new vaccines. Importantly, our study carries implications regarding treatment with anti-CD20 therapies for other conditions, such as rheumatological or neoplastic disorders, and potentially regarding responses to other viruses beyond SARS-CoV-2.
Footnotes
Author Contributions: S.K. contributed conception and design of the study, acquisition and analysis of data, interpretation of results, and writing of manuscript. R.S.F. contributed to conception and design of the study, acquisition and analysis of data, interpretation of results, and writing of manuscript. T.D. contributed to acquisition and analysis of data, interpretation of results, and critical review of manuscript. T.L. contributed to interpretation and analysis of results, statistical analysis, and critical review of manuscript. T.S. contributed to acquisition and analysis of data, interpretation of results, and critical review of manuscript. R.B. contributed to acquisition of data and critical review of manuscript. L.Z.-R. contributed to acquisition and analysis of data, interpretation of results, and critical review of manuscript. A.H. contributed to conception and design of the study, acquisition and analysis of data, interpretation of results, and writing of manuscript.
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: S.K. reports personal fees from Biogen, Alexion Pharmaceuticals, and EMD Serono, as well as grant funding from Biogen. R.S.F. reports personal fees from Alexion Pharmaceuticals and Genentech as well as grant funding from Novartis, Biogen, and The National MS Society outside the submitted work. T.D. reports personal fees from Biogen Idec and Alexion Pharmaceuticals. T.L. has nothing to disclose. T.S. reports clinical fellowship funding from the National Multiple Sclerosis Society, Biogen, honoraria from the American Academy of Neurology, and research grants to institution from PCORI and Consortium for MS Centers outside the submitted work. R.B. reports personal fees from Biogen Idec and Sanofi Genzyme. L.Z.-R. reports personal fees from Biogen, Genentech and Novartis for work as scientific advisor and research grants to institution from Biogen, Genentech, and Consortium for MS Centers outside the submitted work. A.H. reports personal fees from Teva, Biogen, Alexion, Horizon, and Banner Life Sciences, as well as research grants from Biogen, the National MS Society, and the Consortium for MS Centers outside the submitted work.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
ORCID iD: Sylvia Klineova 
https://orcid.org/0000-0002-0406-0460
Contributor Information
Sylvia Klineova, The Corinne Goldsmith Dickinson Center for MS at Icahn School of Medicine at Mount Sinai Hospital, New York, NY, USA.
Rebecca Straus Farber, Multiple Sclerosis Clinical and Research Center, Columbia University Irving Medical Center, New York Presbyterian, New York, NY, USA.
Tracy DeAngelis, Neurological Associates of Long Island, New Hyde Park, NY, USA.
Tungming Leung, Biostatistics Unit, Feinstein Institutes for Medical Research, Northwell Health, Great Neck, NY, USA.
Tyler Smith, NYU Multiple Sclerosis Comprehensive Care Center, NYU Langone Health, New York, NY, USA.
Richard Blanck, Neurological Associates of Long Island, New Hyde Park, NY, USA.
Lana Zhovtis-Ryerson, NYU Multiple Sclerosis Comprehensive Care Center, NYU Langone Health, New York, NY, USA.
Asaff Harel, Northwell Comprehensive Multiple Sclerosis Center, Lenox Hill Hospital and North Shore University Hospital, New York, NY, USA.
References
- 1.Klineova S, Harel A, Straus Farber R, et al. Outcomes of COVID-19 infection in multiple sclerosis and related conditions: One-year pandemic experience of the multicenter New York COVID-19 neuroimmunology consortium (NYCNIC). Mult Scler Relat Disord 2021; 55: 103153. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Louapre C, Collongues N, Stankoff B, et al. Clinical characteristics and outcomes in patients with coronavirus disease 2019 and multiple sclerosis. JAMA Neurol 2020; 77(9): 1079–1088. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Parrotta E, Kister I, Charvet L, et al. COVID-19 outcomes in MS: Observational study of early experience from NYU multiple sclerosis comprehensive care center. Neurol Neuroimmunol Neuroinflamm 2020; 7(5): e835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Salter A, Fox RJ, Newsome SD, et al. Outcomes and risk factors associated with SARS-CoV-2 infection in a North American registry of patients with multiple sclerosis. JAMA Neurol 2021; 78: 699–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sormani MP, De Rossi N, Schiavetti I, et al. Disease-modifying therapies and coronavirus disease 2019 severity in multiple sclerosis. Ann Neurol 2021; 89(4): 780–789. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Ibarrondo FJ, Fulcher JA, Goodman-Meza D, et al. Rapid decay of anti–SARS-CoV-2 antibodies in persons with mild COVID-19. New Engl J Med 2020; 383(11): 1085–1087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Maillart E, Papeix C, Lubetzki C, et al. Beyond COVID-19: DO MS/NMO-SD patients treated with anti-CD20 therapies develop SARS-CoV2 antibodies? Mult Scler Relat Disord 2020; 46: 102482. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Thornton JR, Harel A. Negative SARS-CoV-2 antibody testing following COVID-19 infection in two MS patients treated with ocrelizumab. Mult Scler Relat Disord 2020; 44: 102341. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Zabalza A, Cárdenas-Robledo S, Tagliani P, et al. COVID-19 in multiple sclerosis patients: Susceptibility, severity risk factors and serological response. Eur J Neurol 2021; 28: 3384–3395. [DOI] [PubMed] [Google Scholar]
- 10.Achiron A, Mandel M, Dreyer-Alster S, et al. Humoral immune response to COVID-19 mRNA vaccine in patients with multiple sclerosis treated with high-efficacy disease-modifying therapies. Ther Adv Neurol Disord 2021; 14: 1012835. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Apostolidis SA, Kakara M, Painter MM, et al. Cellular and humoral immune responses following SARS-CoV-2 mRNA vaccination in patients with multiple sclerosis on anti-CD20 therapy. Nat Med 2021; 27(11): 1990–2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Brill L, Rechtman A, Zveik O, et al. Humoral and T-cell response to SARS-CoV-2 vaccination in patients with multiple sclerosis treated with ocrelizumab. JAMA Neurol 2021; 78(12): 1510–1514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bsteh G, Dürauer S, Assar H, et al. Humoral immune response after COVID-19 in multiple sclerosis: A nation-wide Austrian study. Mult Scler 2021; 27(14): 2209–2218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sand IK, Gnjatic S, Krammer F, et al. Evaluation of immunological responses to third COVID-19 vaccine among people treated with sphingosine receptor-1 modulators and anti-CD20 therapy. Epub ahead of print 14 June 2022. DOI: 10.1101/2022.06.10.22276253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Satyanarayan S, Safi N, Sorets T, et al. Differential antibody response to COVID-19 vaccines across immunomodulatory therapies for multiple sclerosis. Mult Scler Relat Disord 2022; 62: 103737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sormani MP, Inglese M, Schiavetti I, et al. Effect of SARS-CoV-2 mRNA vaccination in MS patients treated with disease modifying therapies. eBioMedicine 2021; 72: 103581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Baker D, Roberts CAK, Pryce G, et al. COVID-19 vaccine-readiness for anti-CD20-depleting therapy in autoimmune diseases. Clin Exp Immunol 2020; 202(2): 149–161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bar-Or A, Calkwood JC, Chognot C, et al. Effect of ocrelizumab on vaccine responses in patients with multiple sclerosis: The VELOCE study. Neurology 2020; 95(14): e1999–e2008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ciotti JR, Valtcheva MV, Cross AH. Effects of MS disease-modifying therapies on responses to vaccinations: A review. Mult Scler Relat Disord 2020; 45: 102439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Stokmaier D, Winthrop K, Chognot C, et al. Effect of ocrelizumab on vaccine responses in patients with multiple sclerosis. Neurology 2018; 90(Suppl. 15): S36.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Bajwa HM, Novak F, Nilsson AC, et al. Persistently reduced humoral and sustained cellular immune response from first to third SARS-CoV-2 mRNA vaccination in anti-CD20-treated multiple sclerosis patients. Mult Scler Relat Disord 2022; 60: 103729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Katz JD, Bouley AJ, Jungquist RM, et al. Humoral and T-cell responses to SARS-CoV-2 vaccination in multiple sclerosis patients treated with ocrelizumab. Mult Scler Relat Disord 2022; 57: 103382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Louapre C, Ibrahim M, Maillart E, et al. Anti-CD20 therapies decrease humoral immune response to SARS-CoV-2 in patients with multiple sclerosis or neuromyelitis optica spectrum disorders. J Neurol Neurosurg Psychiatry 2022; 93(1): 24–31. [DOI] [PubMed] [Google Scholar]
- 24.Madelon N, Heikkila N, Sabater Royo I, et al. Omicron-specific cytotoxic T-cell responses after a third dose of mRNA COVID-19 vaccine among patients with multiple sclerosis treated with ocrelizumab. JAMA Neurol 2022; 79(4): 399–404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Meyer-Arndt L, Braun J, Fauchere F, et al. SARS-CoV-2 mRNA vaccinations fail to elicit humoral and cellular immune responses in patients with multiple sclerosis receiving fingolimod. J Neurol Neurosurg Psychiatry 2022; 93(9): 960–971. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Palomares Cabeza V, Kummer LYL, Wieske L, et al. Longitudinal T-cell responses after a third SARS-CoV-2 vaccination in patients with multiple sclerosis on ocrelizumab or fingolimod. Neurol Neuroimmunol Neuroinflamm 2022; 9(4): e1178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Tortorella C, Aiello A, Gasperini C, et al. Humoral- and T-cell-specific immune responses to SARS-CoV-2 mRNA vaccination in patients with MS using different disease-modifying therapies. Neurology 2022; 98(5): e541–e554. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Madelon N, Lauper K, Breville G, et al. Robust T-cell responses in anti-CD20-treated patients following COVID-19 vaccination: A prospective cohort study. Clin Infect Dis 2022; 75(1): e1037–e1045. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Elecsys anti-SARS-CoV-2, https://www.fda.gov/media/137605/download
- 30.COVID-19 booster vaccine in autoimmune disease non-responders, https://clinicaltrials.gov/ct2/show/NCT05000216
- 31.COVID-19 protection after transplant pilot study (CPAT), https://clinicaltrials.gov/ct2/show/NCT04969263
- 32.JAMA Network. Tixagevimab and cilgavimab (Evusheld) for pre-exposure prophylaxis of COVID-19. JAMA 2022; 327(4): 384–385. [DOI] [PubMed] [Google Scholar]
- 33.Kageyama T, Ikeda K, Tanaka S, et al. Antibody responses to BNT162b2 mRNA COVID-19 vaccine and their predictors among healthcare workers in a tertiary referral hospital in Japan. Clin Microbiol Infect 2021; 27(12): 1861.e1–1861.e5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sahin U, Muik A, Vogler I, et al. BNT162b2 vaccine induces neutralizing antibodies and poly-specific T cells in humans. Nature 2021; 595(7868): 572–577. [DOI] [PubMed] [Google Scholar]
- 35.Disanto G, Galante A, Cantu’ M, et al. Longitudinal postvaccine SARS-CoV-2 immunoglobulin G titers, memory B-cell responses, and risk of COVID-19 in multiple sclerosis over 1 year. Neurol Neuroimmunol Neuroinflamm 2023; 10(1): e200043. [DOI] [PMC free article] [PubMed] [Google Scholar]