Nationwide Italian registry reveals who responds best to rituximab in NMOSD and MOGAD

Two rare autoimmune diseases of the central nervous system, once lumped together with multiple sclerosis, are finally getting the large-scale statistical scrutiny they deserve. Neuromyelitis optica spectrum disorder (NMOSD) driven by antibodies against aquaporin-4 (AQP4) and myelin oligodendrocyte glycoprotein antibody-associated disease (MOGAD) share overlapping symptoms—optic neuritis, transverse myelitis, devastating relapses—but they differ profoundly in biology, demographics, and, as a new Italian study shows, in how they respond to treatment. A nationwide analysis of 664 patients, drawn from 97 centres across Italy and published in the Journal of Neurology, offers one of the clearest real-world pictures yet of who relapses, who stays well, and which drugs actually deliver.

The study, led by Carla Tortorella and colleagues on behalf of the Italian Multiple Sclerosis and Related Disorders (I-MS&RD) Registry, is notable for its sheer scale. Rare diseases are notoriously difficult to study: individual centres may see only a handful of cases per year, making it nearly impossible to draw statistically meaningful conclusions. By pooling data from nearly a hundred centres, the Italian team assembled 406 patients with AQP4-positive NMOSD and 258 with MOGAD—a cohort large enough to disentangle the clinical fingerprints of the two conditions and to test which factors predict treatment success.

The demographic contrasts between the two diseases were striking. Female sex predominated in AQP4-positive NMOSD, consistent with its well-documented skew toward women, whereas MOGAD was significantly more likely to begin in childhood. Patients with AQP4-positive NMOSD also carried greater disability at diagnosis than those with MOGAD, a finding that echoes the aggressive, complement-driven attack on astrocytes that characterises AQP4 antibody disease. In MOGAD, by contrast, the antibody targets a myelin surface protein, and the resulting inflammation is often more transient, with many patients experiencing monophasic or self-limiting courses.

When it came to treatment, one drug stood out: rituximab, a monoclonal antibody that depletes CD20-positive B cells, was the most frequently prescribed therapy across both groups. This reflects clinical reality in much of Europe, where rituximab has long served as the workhorse off-label option for antibody-mediated CNS disease, even as newer approved agents—eculizumab, satralizumab, inebilizumab—have entered the market. The Italian registry data thus capture a treatment landscape in transition, where cost, availability, and physician familiarity still shape prescribing patterns as much as trial evidence does.

The central question the researchers posed was deceptively simple: does treatment class matter for keeping patients relapse-free? Using time-to-first-relapse analysis, they compared monoclonal antibodies against other immunotherapies. In AQP4-positive NMOSD, monoclonal antibody treatment was associated with significantly longer relapse-free survival, with a log-rank p-value of 0.021—a statistically robust signal. In MOGAD, however, the same comparison showed no significant advantage. This asymmetry is biologically plausible: AQP4 antibodies are produced by plasmablasts dependent on a continuously replenished B-cell pool, making B-cell depletion or upstream immunological blockade particularly effective. MOGAD’s relapse biology is less clearly B-cell driven, and many patients do well even on conventional agents.

Drilling deeper into the rituximab-treated subgroup revealed predictors of failure that differed sharply between the two diseases. Among AQP4-positive NMOSD patients on rituximab, male sex and myelitis at disease onset independently predicted a lower likelihood of remaining relapse-free. Male sex carried an odds ratio of 0.37 (95% confidence interval 0.14–0.94; p = 0.038), and spinal cord involvement at onset an odds ratio of 0.36 (95% CI 0.15–0.83; p = 0.017). In other words, men and patients presenting with transverse myelitis were roughly two-and-a-half times less likely to stay relapse-free on rituximab than their counterparts. The sex finding is intriguing and may relate to pharmacokinetic differences—rituximab clearance varies with body weight and sex—or to underlying immunological differences in antibody production.

For MOGAD patients treated with rituximab, the picture was entirely different. Optic neuritis at onset was the only independent predictor of relapse-free status at follow-up, with an odds ratio of 0.20 (95% CI 0.04–0.90; p = 0.038). Patients whose disease began with inflammation of the optic nerve were five times less likely to remain relapse-free on rituximab than those with other presentations. This aligns with a growing literature suggesting that MOGAD with optic nerve involvement carries a higher relapse risk, and it raises the practical question of whether such patients might benefit from escalation to other agents or closer monitoring of B-cell reconstitution.

The study’s methodology deserves attention for what it represents. The I-MS&RD Registry is a newly established national infrastructure, and this analysis is among its first major outputs. Multivariable logistic regression was used to adjust for confounders, and the authors were careful to frame their findings as real-world evidence rather than causal proof—treatment assignment in observational registries is never random, and healthier or more severe patients may be systematically steered toward particular drugs. Still, the consistency of the monoclonal antibody signal in AQP4-positive NMOSD with randomised trial data for agents like eculizumab and satralizumab lends credibility to the registry’s approach.

The findings arrive at a moment when the field is converging on a consensus that NMOSD and MOGAD must be treated as distinct entities. International diagnostic criteria published in 2015 for NMOSD and in 2023 for MOGAD have formalised this separation, and trials such as N-MOmentum and the satralizumab phase 3 studies have built the evidence base for targeted therapy in AQP4-positive disease. What has been missing is granular, population-level data on how these diseases behave outside trial settings—particularly for MOGAD, which lacks any approved disease-specific therapy and is managed largely with off-label immunosuppression.

For clinicians, the practical message is that phenotype should guide treatment intensity. A man with AQP4-positive NMOSD presenting with myelitis may need more than rituximab alone, or more frequent monitoring of B-cell counts and antibody titres. A MOGAD patient whose first attack is optic neuritis warrants heightened vigilance for relapse even on B-cell depletion. For researchers, the Italian registry provides a template for how rare-disease networks can generate actionable evidence: large numbers, standardised data collection, and analysis frameworks that respect the biological distinctions between closely related conditions. As more countries build similar registries, the era of guessing how to treat these diseases may finally be drawing to a close.

Subject of Research: Clinical phenotype and treatment response in AQP4-IgG-positive NMOSD and MOGAD: insights from the Italian I-MS&RD registry

Article Title: Clinical phenotype and treatment response in AQP4-IgG-positive NMOSD and MOGAD: insights from the Italian I-MS&RD registry

Article References: Tortorella, C., Cruciani, A., Patti, F., Capobianco, M., Iaffaldano, P., Mirabella, M., Inglese, M., Pastò, L., Haggiag, S., Prosperini, L., Ruggieri, S., Fantozzi, R., Foschi, M., De Luca, G., Vianello, M., Salemi, G., Montepietra, S., Ferraro, D., Cavalla, P., … Lepore, V. (2026). Clinical phenotype and treatment response in AQP4-IgG-positive NMOSD and MOGAD: insights from the Italian I-MS&RD registry. Journal of Neurology, 273(11), Article 665. https://doi.org/10.1007/s00415-026-14185-2

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14185-2

Keywords: Clinical, phenotype, treatment, response, AQP4-IgG-positive, NMOSD, MOGAD, insights, Italian, I-MS, registry, scientific research