ABSTRACT
Introduction: Some multiple sclerosis (MS) disease-modifying therapies (DMTs) are associated with blunted humoral vaccination responses, but relevance for SARS-CoV-2 infection is unclear. Objective(s): To determine SARS-CoV-2 exposure rates and formation of antibody memory among participants of the COMparison Between All immunoTherapies for MS (COMBAT-MS;NCT03193866) and the Immunomodulation and MS Epidemiology (IMSE) studies. Aim(s): To determine SARS-CoV2 serological response of people living with MS (pwMS). Method(s): Using a multiplex bead-based assay we determined SARS-CoV-2 spike and nucleocapsid antibody levels in 3,723 pwMS in paired serum samples (n=7,157) donated prior (<January 31st 2020) and during the pandemic (July-October 2020);16.6% had natalizumab, 6.4% fingolimod, 9.7% dimethyl fumarate, 1.9% interferon beta, 50.4% rituximab, 1.4% cladribine, 7.6% other DMTs, and 6.1% were untreated. Median fluorescent intensity (MFI) and bead-count were determined for spike and nucleocapsid antibodies, and samples were regarded as positive only when reactive to both viral antigens. Hazard ratios, from multivariable Cox regression models, were derived to assess association between antibody levels above cut-off for each antigen, comparing exposure to rituximab or fingolimod at time of sampling vs. other reference DMTs. All models were adjusted for age, sex, treatment center, time since reported infection, MS severity, disease duration, and number of previous DMTs. Result(s): Specificity and sensitivity of the assay for SARS-CoV-2 was 100% and 99.7%, respectively. The proportion of positive samples for SARS-CoV-2 differed moderately across DMTs with the highest values among cladribine-treated (7.4%) and the lowest number among rituximab-treated pwMS (3.9%). Similarly, the proportion of positive cases not reported in the Swedish MS registry varied from 100% for cladribine to 33.3% among untreated pwMS. Comparing levels of antibodies titers showed that levels were lower among those treated with rituximab or fingolimod vs interferon treated pwMS. Point estimates indicated a similar trend comparing rituximab or fingolimod vs untreated pwMS. Conclusion(s): Overall rates of SARS-CoV-2 antibody positivity after the first COVID-19 wave differed only moderately across DMTs, while antibody levels were lower with rituximab or fingolimod compared to interferon-treated pwMS. This indicates quantitative rather than qualitative differences in the humoral response to infection.
ABSTRACT
Introduction: Mounting evidence suggests that B-cell depleting disease-modifying therapies (DMTs) may be associated with an increased risk of a moderate to severe COVID-19. Use of rituximab (RTX) is frequent in Sweden, a country that suffered high COVID-19 rates. Objective: To explore the association between timing and dose of RTX and risk of hospitalization for COVID-19 in a well-characterized cohort of MS patients. Methods: We included participants in COMBAT-MS, (NCT03193866) with ongoing treatment with RTX. Data include detailed demographic and clinical information, DMT exposure (date and dose of all RTX infusions), COVID-19 information (dates of infection and hospitalization as reported by the patient), as well as patient-reported outcomes. Among all persons who developed COVID-19, we measured the odds of hospitalization for COVID-19 as they related to 1) the time between most recent RTX infusion and COVID-19 onset date (measured in months), and 2) the total lifetime cumulative dose of RTX received (measured in grams), using logistic regression. The reference cohort included all persons with 'mild' COVID-19, not requiring hospitalization. Models were adjusted for age and sex with results reported as odds ratios (OR) and 95% confidence intervals (CI). Results: Of 3391 persons actively enrolled in the COMBAT-MS cohort, 326 (9.6%) developed COVID-19 between March 1, 2020 and April 30, 2021. Amongst these patients, 172 (52.8%) were exposed to RTX prior to their COVID-19 onset, and 26 (15.1%) required hospitalization. No deaths occurred. Median time between last RTX infusion and COVID-19 onset was 6.1 months [interquartile range (IQR): 3.9, 11.0] among mild cases, and 4.6 months [IQR: 3.6, 5.6] among those hospitalized for COVID-19 (difference not significant, Mann-Whitney U-test p=0.16). Persons with mild COVID-19 had a median cumulative lifetime RTX dose of 3.5g [IQR: 2.5, 4.5], compared to 3.3g [IQR:2.6, 4.5] for hospitalized cases. The time between most recent DMT infusion and COVID-19 onset was not associated with altered odds of requiring hospitalization (age-and sexadjusted OR: 0.97 (95%CI: 0.90-1.03), nor was lifetime cumulative dose of RTX (age and sex-adjusted OR: 1.05 (95%CI: 0.83-1.31). Conclusions: Among COMBAT-MS participants who contracted COVID-19, there was no significant association between timing of RTX infusion nor cumulative lifetime RTX dose and the odds of hospitalization for COVID-19.