NMOSD Treatment: Separating MOGAD and Choosing Among Biologics

The Clinical Problem in NMOSD Treatment

Neuromyelitis optica spectrum disorder is a relapsing autoimmune disease of the optic nerves, spinal cord, and brainstem. Each attack can leave fixed deficits, so relapse prevention drives the whole approach to NMOSD treatment.

Roughly two-thirds of patients carry AQP4-IgG, an antibody that marks complement-mediated astrocyte injury. The 2015 International Panel for NMO Diagnosis criteria formalised this serologically stratified definition and separated AQP4-IgG-positive from seronegative disease.

The central practical question has two parts. First, which patients actually have AQP4-IgG-positive NMOSD as opposed to a mimic. Second, once confirmed, which long-term immunotherapy to start — because the modern options differ sharply in target, route, and monitoring.

Separating MOGAD from AQP4-IgG disease

The most consequential mimic is MOGAD. Once labelled “seronegative NMOSD,” it is now recognised as a distinct disease driven by antibodies to myelin oligodendrocyte glycoprotein. The 2023 International MOGAD Panel proposed criteria that make MOG-IgG, tested on a validated cell-based assay, a core requirement.

The distinction is not academic. MOGAD tends toward better attack recovery, is often monophasic in children, and carries a lower long-term relapse risk than AQP4-IgG-positive NMOSD. These differences change whether, when, and how to treat.

Crucially, every pivotal biologic trial enrolled AQP4-IgG-positive patients, not MOGAD. Applying NMOSD trial data to a MOG-antibody patient assumes an equivalence the evidence never tested. Serostatus therefore has to be settled before the treatment conversation begins.

Available Approaches in NMOSD Treatment

For AQP4-IgG-positive disease, four biologics reached regulatory approval on the strength of phase 3 trials, and a fifth agent — rituximab — is supported by a randomised trial and extensive real-world use. They fall into three mechanistic families.

Complement inhibition — eculizumab and ravulizumab

Complement inhibitors target the terminal cascade that drives AQP4-antibody injury. In the PREVENT trial, eculizumab led to adjudicated relapse in 3% of patients versus 43% on placebo, a hazard ratio of 0.06 (95% CI 0.02–0.20). This was the first randomised trial to show clear benefit in NMOSD.

Ravulizumab binds the same C5 epitope but has a longer half-life, allowing eight-weekly rather than two-weekly dosing. In the open-label CHAMPION-NMOSD study, no ravulizumab-treated patient had an adjudicated relapse, against an external placebo comparator drawn from PREVENT.

Two design caveats matter. CHAMPION-NMOSD used an external, non-randomised control, so its effect estimate is less robust than a blinded trial. And complement inhibition raises meningococcal infection risk, mandating vaccination before treatment.

Interleukin-6 receptor blockade — satralizumab

Satralizumab blocks IL-6 signalling, implicated in antibody production and blood–brain-barrier disruption. Two blinded phase 3 trials tested it: SAkuraSky as an add-on to background immunosuppression, and SAkuraStar as monotherapy.

Benefit concentrated in AQP4-IgG-positive patients. In SAkuraSky, relapse occurred in 11% on satralizumab versus 43% on placebo among seropositive patients (HR 0.21, 95% CI 0.06–0.75). SAkuraStar showed a comparable seropositive hazard ratio of 0.26 (95% CI 0.11–0.63).

Because IL-6 blockade suppresses C-reactive protein and can blunt fever, clinicians lose some conventional signals of infection — a monitoring consideration rather than a contraindication.

B-cell depletion — inebilizumab and rituximab

B-cell depletion removes the antibody-producing lineage. Inebilizumab targets CD19, depleting a broader B-cell range than anti-CD20 agents. In N-MOmentum, attacks occurred in 12% on inebilizumab versus 39% on placebo (HR 0.27, 95% CI 0.15–0.50).

Rituximab, an anti-CD20 antibody, is not formally approved for NMOSD in most regions but is used extensively. The randomised RIN-1 trial recorded no relapses on rituximab versus seven of nineteen (37%) on placebo over 72 weeks, with the group difference of 36.8% (95% CI 12.3–65.5).

RIN-1 was small and enrolled patients with relatively mild activity, so its precision is limited. Even so, its result aligns with large observational cohorts, which is why rituximab remains a common first-line choice where approved agents are inaccessible.

Comparative Analysis

The decisive limitation is structural: there are no head-to-head randomised trials among these agents. Each was compared with placebo, in different populations, under different background-therapy rules. That makes direct ranking impossible from trial data alone.

Indirect network meta-analyses that pool the pivotal trials have suggested numerically greater relapse-risk reduction with the complement inhibitors than with the B-cell- or IL-6-targeted agents. These comparisons carry wide credible intervals and are confounded by trial design, so they inform intuition rather than settle it.

Several cross-trial differences deserve weight. PREVENT permitted continued background immunosuppression, whereas SAkuraStar tested monotherapy; the placebo groups therefore differ. Effect sizes cannot be lifted between trials without accounting for these mismatches.

In practice, the choice among AQP4-IgG-positive options often rests on non-efficacy factors: infusion versus subcutaneous route, dosing interval, meningococcal risk with complement inhibitors, infection and immunoglobulin monitoring with B-cell depletion, and blunted inflammatory markers with IL-6 blockade.

Why MOGAD diverges

The clearest evidence that serostatus should steer NMOSD treatment comes from rituximab. In a cohort of 111 NMOSD and 23 MOGAD patients, the risk of relapse while on rituximab was 2.8-fold higher in MOGAD (HR 2.8, 95% CI 1.5–5.2). Relapses on treatment occurred in 28% of NMOSD versus 61% of MOGAD patients.

This is observational, so it shows association rather than a causal treatment failure. But it signals that B-cell depletion, so effective in AQP4-IgG-positive disease, is a less reliable strategy in MOGAD. Applying an NMOSD playbook to a MOG-antibody patient can therefore under-treat.

Quality and Consistency of the Evidence

Evidence strength varies sharply by question. The colour ratings below reflect study design, consistency across sources, and precision.

• 🟢High — relapse prevention in AQP4-IgG+ NMOSD with approved biologics. Multiple concordant, blinded phase 3 randomised trials with large effect sizes.
• 🟢High — rituximab efficacy in AQP4-IgG+ NMOSD. A randomised trial (RIN-1) reinforced by consistent real-world cohorts.
• 🟡Moderate — comparative ranking among biologics. Only indirect network meta-analyses; confounded by design, with no head-to-head confirmation.
• 🟡Low-to-moderate — seronegative NMOSD efficacy. Subgroup estimates are imprecise and confidence intervals cross 1 (e.g. satralizumab seronegative HR 0.66, 95% CI 0.20–2.24).
• 🔴Low / emerging — MOGAD relapse prevention. One conference-reported phase 3 trial plus observational data; no approved therapy at the time of writing.

What the Evidence Does Not Show

It does not show which biologic is superior. Without randomised head-to-head trials, apparent differences in relapse-risk reduction cannot be attributed to the drugs rather than to divergent trial populations and control arms.

It does not establish efficacy in AQP4-IgG-seronegative NMOSD. The pivotal benefits sit in seropositive subgroups; seronegative estimates are underpowered, and confidence intervals often include no effect.

It does not define optimal treatment duration or safe de-escalation. Whether, when, and how to taper complement inhibition or B-cell depletion after prolonged remission remains untested in randomised settings.

It does not yet confirm an approved MOGAD therapy. The satralizumab MOGAD result is promising but conference-level; the optimal agent, treatment threshold, and duration in MOGAD are unresolved.

It does not compare long-term safety head-to-head. Meningococcal risk, serious infection, and hypogammaglobulinaemia differ by mechanism, but no trial ranks the agents on these harms directly.

Practical Implications

These points describe what the current evidence supports; they are not prescriptive directives and do not replace individualised assessment.

Confirm serostatus first. Test AQP4-IgG and MOG-IgG with validated cell-based assays before committing to long-term NMOSD treatment, because trial evidence and expected response depend on which antibody, if any, is present.

For AQP4-IgG-positive NMOSD, the evidence supports starting relapse prevention after the first attack. The four approved biologics carry phase 3 support; rituximab is a widely used, RCT-supported off-label option where access or cost constrains choice.

Match the agent to mechanism-specific logistics. Complement inhibitors require meningococcal vaccination; B-cell-depleting agents warrant infection and immunoglobulin monitoring; IL-6 blockade can mask inflammatory markers.

For MOGAD, preventive therapy is generally reserved for relapsing disease rather than a single event. Rituximab appears less reliable than in NMOSD, and emerging IL-6-directed data should be read cautiously until peer-reviewed and, where relevant, approved.

Verify all drug dosing against current prescribing information, since regimens, monitoring, and approved indications differ across agents and regions.

Evidence Grade and Bottom Line