, according to a new analysis of an Italian cohort of patients with MS. The study confirmed that steroid exposure in the month before COVID-19 symptom onset is tied to more severe disease, and anti-CD20 therapy poses similar risks. But the researchers noted that interferon and possibly teriflunomide were associated with a protective effect in the multivariate analysis.
Maria Pia Sormani, PhD, who is a professor of biostatistics at the University of Genoa, presented the study at the 2021 annual meeting of the American Academy of Neurology.
The results confirm some previous analyses, and add to the body of evidence clinicians rely on, according to Jiwon Oh, MD, PhD, who moderated the session. “These data about the risk with the anti-CD20 therapies have been around for a while, but it seems that risk is pretty apparent, with this registry and other registries around the world. It affects counseling to patients on anti-CD20 therapies. We would counsel them to be cautious, obviously, follow public health precautions, but maybe be even more cautious. It affects our recommendations about the urgency of vaccination in these folks, how high priority they should be,” Dr. Oh said in an interview. She is the clinical director of the Barlo MS Center at St. Michael’s Unity Health in Toronto.
The analysis also hinted at complexities within demographics that might help explain some of the differing outcomes of infections. “We have learned that the course of the viral infection per se may not be the cause of severe outcomes, but the exaggerated inflammatory response to the virus is mainly responsible for intubations and deaths. The hypothesis we are investigating is whether anti-CD20 therapies can cause a more severe viral infection (that is something already known for other viral infections) but do not play a crucial role in causing the explosion of the inflammatory process,” said Dr. Sormani in an email.
The group plans to look at the risk of anti-CD20 therapies in different age groups, “to try to understand the underlying mechanism through which anti-CD20 increases the risk of more severe outcome,” she said.
Dr. Sormani presented an analysis of 3,274 patients with MS who contracted COVID-19 in Italy. The mean age was 44, the median Expanded Disability Status Scale (EDSS) score was 2, Among the study cohort, 68.6% were female; 14% had progressive MS and 26 patients died. Patients who died had a mean age of 63, 48% were female, 73% had progressive MS, and 50% were not on any DMT.
The researchers used ordinal logistic regression that “orders” outcome on a severity scale of 0 (mild disease, no pneumonia or hospitalization), 1 (pneumonia or hospitalization, n = 184), or 2 (ICU admission or death, n = 36). They calculated the odds ratio of moving from 0 to 1, or 1 to 2, and carried the assumption that the risk is the same. For example, an odds ratio of 2 for males versus females would mean that males are twice as likely to be hospitalized and twice as likely to go from being hospitalized to going to the ICU or dying.
The researchers found that older age, male sex, and comorbidities increase risk of worse COVID-19 outcomes. Exposure to methylprednisolone 1 month before COVID-19 symptom onset carried an increased risk (OR, 2.33; P = .03). Compared with no therapy, receiving interferon was associated with lower risk (OR, 0.34; P = .009) and teriflunomide trended towards an association with better outcomes (OR, 0.49; P = .054). Anti-CD20 treatment (ocrelizumab or rituximab) was linked to worse outcomes (OR, 1.89; P = .012) overall, which held up when ocrelizumab (OR, 1.71; P = .04) and rituximab (OR, 2.77; P = .03) were considered separately.
To understand why the risk of ocrelizumab might be lower, the researchers examined risk by duration of anti-CD20 treatment, and found that risk increased with increased duration of treatment, with the lowest risk at treatment duration less than 6 months (OR, 1.56; 95% CI, 0.65-3.77; not significant), followed by 6 months to 1 year (OR, 1.68; 95% CI, 0.69-4.03; P < .001), 1-2 years (OR, 1.74; 95% CI, 0.83-3.64; trend), and the highest risk at more than 2 years (OR, 2.75; 95% CI, 1.28-5.88).
Dr. Sormani suggested that the greater risk associated with rituximab may be because of a tendency towards longer treatment length, since patients treated with rituximab were more often treated for greater lengths of time; 11% had been treated for 6 months or less (vs. 24% of ocrelizumab patients); 26%, 6-12 months (vs. 18% ocrelizumab); 19%, 1-2 years (vs. 37% ocrelizumab); and 44%, 2 years or longer (vs. 21% ocrelizumab).
Dr. Sormani has received consulting fees from Biogen, GeNeuro, Genzyme, MedDay, Merck KGaA, Novartis, Roche, and Immunic. The platform for data collection was donated by Merck. Dr. Oh has consulted for Roche, Celgene, Biogen-Idec, EMD-Serono, Sanofi-Genzyme, Novartis, Alexion. She has been on a scientific advisory or data safety monitoring board for Roche, Biogen-Idec, and Sanofi-Genzyme.