← Back
NeuroTrials.ai
Neurology Clinical Trial Database

NOMADMUS

French NOMADMUS Cohort Overview: Landscape Evolution of AQP4+NMOSD and MOGAD From 2010 to 2024

Year of Publication: 2026

Authors: Thomas Roux, François-Xavier Lejeune, Romain Casey, ..., Elisabeth Maillart; NOMADMUS Study Group

Journal: Neurology

Citation: Roux T, Lejeune FX, Casey R, et al. French NOMADMUS Cohort Overview: Landscape Evolution of AQP4+NMOSD and MOGAD From 2010 to 2024. Neurology. 2026;106(12):e218073. doi:10.1212/WNL.0000000000218073


Clinical Question

How have the diagnosis, treatment, disease activity, and disability of AQP4+NMOSD and MOGAD evolved in France over 15 years?

Bottom Line

Over 15 years, the French NOMADMUS registry shows shortened time to diagnosis, markedly reduced relapse activity, and a shift toward rituximab and newer biologics in AQP4+NMOSD; rituximab exposure was associated with reduced disability accrual, and MOGAD carried a substantially better long-term disability prognosis than AQP4+NMOSD.

Major Points

  • Nationwide French registry of 1,726 patients (769 AQP4+NMOSD, 957 MOGAD) across 98 centers, data extracted June 8, 2024.
  • Mean annualized relapse rate declined dramatically from 2010-2014 to 2022-2023: AQP4+NMOSD 0.45 (0.41-0.48) to 0.04 (0.03-0.05); MOGAD 0.35 (0.31-0.39) to 0.10 (0.08-0.12); both p<0.0001.
  • Therapeutic shift in AQP4+NMOSD: azathioprine/mycophenolate fell 51.6% to 40.3% while rituximab rose 32.6% to 43.6%, with emergence of new agents (eculizumab, tocilizumab, satralizumab, inebilizumab) reaching 12.5% in 2022-2023.
  • Rituximab exposure was associated with lower risk of reaching EDSS 6 in AQP4+NMOSD (HR 0.38; 95% CI 0.21-0.68; p<0.001); older age at onset increased that risk (HR 1.04; p<0.001). Neither factor influenced MOGAD.
  • MOGAD had a better disability trajectory than AQP4+NMOSD: EDSS 3 at 5 years 2.8% vs 20%; EDSS 6 at 5 years 0.94% vs 7.4%.
  • Methylprednisolone was used in nearly all acute relapses (97.4% AQP4+NMOSD, 97.6% MOGAD); plasma exchange use increased over time, especially in AQP4+NMOSD (28.7% to 54.5%).
  • Time from onset to first positive antibody test and to first immunosuppressive treatment shortened over the study period in both groups.

Design

Study Type: Retrospective nationwide observational registry cohort study

Randomization:

Blinding: None (observational registry)

Enrollment Period: 2010 to 2024 (data extracted June 8, 2024)

Follow-up Duration: Mean 5.1 years overall (AQP4+NMOSD 6.9 y; MOGAD 3.4 y)

Centers: 98

Countries: France

Sample Size: 1726

Analyzed: 1726

Analysis: Welch's t tests for ARR comparisons across periods; Cox proportional hazards models for time to EDSS 3 and EDSS 6 (multivariate)

Registration: NCT02889965


Inclusion Criteria

  • Enrolled in the French NOMADMUS cohort (nested in OFSEP) across 98 centers
  • AQP4-IgG seropositive NMOSD fulfilling 2015 international criteria (pre-2015 cases re-evaluated and reclassified)
  • OR MOG-IgG seropositive MOGAD fulfilling 2023 criteria with a core clinical syndrome and clear positive MOG-IgG (mostly live cell-based assay in Lyon)

Exclusion Criteria

  • Double seronegative NMO-related disorders, including 2015 seronegative NMOSD, isolated LETM, or atypical optic neuritis (n=451)
  • Insufficient data (n=32)

Baseline Characteristics

CharacteristicAQP4+NMOSDMOGADTotal
N7699571726
Female n (%)674 (87.6)509 (53.2)1,183 (68.5)
Age at onset (y)42.0 (17.6)33.8 (18.0)37.4 (18.3)
Onset <18 years (%)53 (6.9)197 (20.6)250 (14.5)
Age at first examination (y)44.9 (16.6)37.9 (16.6)
Follow-up duration (y)6.9 (5.8)3.4 (3.8)
Disease duration (y)9.6 (8.6)5.1 (7.0)
First relapse optic neuritis (%)316 (41.1)612 (63.9)
First relapse myelitis (%)230 (29.9)164 (17.1)
First relapse encephalic/brainstem (%)227 (29.5)216 (22.6)
Positive oligoclonal bands (%)148 (26.0)104 (14.2)

Arms

FieldControlMOGAD
N769957
InterventionAQP4-IgG seropositive NMOSD (reference group); real-world immunosuppressive and acute relapse managementMOG-IgG seropositive MOGAD; real-world immunosuppressive and acute relapse management
DurationMean follow-up 6.9 yearsMean follow-up 3.4 years

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Change in mean annualized relapse rate (ARR) over time (disease activity)PrimaryAQP4+NMOSD: 0.45 (0.41-0.48) in 2010-2014 to 0.04 (0.03-0.05) in 2022-2023MOGAD: 0.35 (0.31-0.39) in 2010-2014 to 0.10 (0.08-0.12) in 2022-2023<0.0001
Secondary0.38<0.001
Secondary1.04<0.001
Secondary
Secondary
Secondary
Secondary
Secondary

Subgroup Analysis

Acute relapse treatment: methylprednisolone used in 97.4% (AQP4+NMOSD) and 97.6% (MOGAD) of relapses; PLEX in 23.1% and 17.1% respectively, increasing over time (AQP4+NMOSD PLEX 28.7% in 2010-2014 to 54.5% in 2022-2023), with increased IVIg use in MOGAD. Untreated patients were rare in AQP4+NMOSD but more frequent in MOGAD.


Criticisms

  • Retrospective, non-randomized registry design limits causal inference (e.g., rituximab-disability association may reflect confounding by indication or era effects).
  • Shorter mean follow-up in MOGAD (3.4 y) than AQP4+NMOSD (6.9 y) may bias disability comparisons.
  • Single-country (France) cohort; generalizability to other health systems uncertain.
  • Declining ARR over time may partly reflect improved ascertainment, earlier diagnosis, and changing case mix rather than treatment effect alone.

Based on: NOMADMUS (Neurology, 2026)

Authors: Thomas Roux, François-Xavier Lejeune, Romain Casey, ..., Elisabeth Maillart; NOMADMUS Study Group

Citation: Roux T, Lejeune FX, Casey R, et al. French NOMADMUS Cohort Overview: Landscape Evolution of AQP4+NMOSD and MOGAD From 2010 to 2024. Neurology. 2026;106(12):e218073. doi:10.1212/WNL.0000000000218073

Content summarized and formatted by NeuroTrials.ai.