Effectiveness of tixagevimab/cilgavimab (Evusheld) in antiCD20‑treated patients with multiple sclerosis and neuromyelitis optica spectrum disorder

Background: AntiCD20 therapy, such as rituximab, ocrelizumab, or ofatumumab, effectively treats patients with multiple sclerosis (pwMS) or neuromyelitis optica spectrum disorder (pwNMOSD) but negatively affects the humoral immune response to COVID-19 vaccination. One strategy to protect these patients is using tixagevimab/cilgavimab (T/C) as pre-exposure prophylaxis. This study aimed to evaluate the effect of T/C on the incidence of COVID-19 in pwMS and pwNMOSD.

Methods: Data in this observational cohort study were collected in two Czech MS centres through ReMuS registry between March 1, 2020 and December 31, 2022. Adult pwMS and pwNMOSD who were (1) treated with antiCD20 therapy at least six months before T/C administration, or at least from February 1, 2022 in the control group; (2) were already on antiCD20 therapy at the time of vaccination or COVID-19 infection; and (3) were on antiCD20 therapy at least 100 days after T/C, or at least 90 days after August 1, 2022 in the control group, were included. Analysis was performed using frequency-based (propensity score matching) and Bayesian statistical methods (informative and non-informative priors).

Results: Using propensity score matching 1:1, 47 patients who received T/C (mean age 45.7 years, median disease duration 12.5 years) were matched with those who did not receive T/C (n = 341; mean age 46.6 years, median disease duration 11.4 years) based on age, MS/NMOSD duration, and number of vaccine doses. None of the T/C patients and three in the control matched group, developed COVID-19 between 10 and 100 days after receiving T/C, August 1, 2022, respectively. The frequency of COVID-19 was not significantly different between groups (p = 0.242). Due to the low number of patients, a Bayesian analysis was also added. Using a non-informative Bayesian prior, the median relative risk of COVID-19 after T/C was 7.6 % (95 % CrI 0.02-115.9 %). The posterior probability of risk difference lower than zero was 96.4 %. Using an informative prior (based on the registration study of Evusheld), the median relative risk of COVID-19 after T/C was 20.2 % (95 % CI 8.4-43.8 %). The posterior probability of the risk difference lower than zero was 100 %.

Conclusion: This work highlights the possible good efficacy of T/C in antiCD20-treated pwMS and pwNMSOD. Based on Bayesian analysis with an informative prior, the T/C group's risk of COVID-19 infection was approximately 20.2 % of the control group's risk. However, given the low frequency of COVID-19, the results of this pilot analysis must be interpreted with caution.