Neuromyelitis optica spectrum disorders (NMOSD) and myelin oligodendrocyte glycoprotein antibody disease (MOGAD) are both immune-mediated demyelinating disorders frequently involving optic nerve and/or spinal cord causing a wide range of neurological disabilities (1). For decades, patients presenting with the clinical picture now linked to NMOSD (i.e., transverse myelitis [TM], unilateral or bilateral optic neuritis [ON], and brain and spinal cord lesions) were often misdiagnosed with severe multiple sclerosis (MS) (2, 3). Later on, the identification and detection of the antibodies to the protein aquaporin-4 (AQP4-IgG) allowed for the differentiation of these patients from those with MS (2). This has fundamentally transformed the diagnosis and pathogenetic understanding of NMOSD, expanding its defined clinical presentation to comprise cases of acute postrema syndrome (APS) (defined as an episode of otherwise interactable unexplained hiccups or nausea and vomiting), brainstem, hypothalamic, diencephalic and cerebral MRI lesions in addition to the originally described optic neuritis and longitudinally extensive spinal cord disease (2).

Similarly, and following years of their association to NMOSD patients with undetectable AQP4-IgG antibodies (4), the characterization of MOG-IgG antibodies—which was initially detected in 2009 in acute disseminated encephalomyelitis (ADEM) cases (5)—has paved the way for establishing MOGAD as a distinct demyelinating disorder, with its diagnosis centered around a confirmed positive MOG-IgG antibodies test, in addition to the presence of a core clinical demyelinating event (e.g., ON, TM, ADEM, cerebral monofocal or polyfocal deficits, brainstem or cerebellar deficits, and cerebral cortical encephalitis, often with seizures) and the exclusion of MS. Similar to NMOSD, a low positive antibodies test requires further clinical or MRI findings, in addition to a seronegative aquaporin AQP4-IgG test (6).

Despite the similarity in clinical features, NMOSD and MOGAD vary in patient demographics, disease course, prognosis, and treatment outcomes. This can be attributed to the pathogenically different immune-mediation pathways, where the main target of immune attack is myelin in MOGAD versus astrocytes in AQP4 + NMOSD (7). Patients with MOGAD also tend to experience more severe acute attacks compared to those with NMOSD (8, 9). However, MOGAD is also associated with better recovery and less irreversible neurological damage, suggesting more favorable long-term outcomes (10). In contrast, NMOSD, specifically in AQP4 seropositive cases, is marked by a higher risk of accumulated disability over time.

Existing MS treatment options were previously employed for NMOSD patients, primarily due to the overlapping clinical pictures of MS and NMOSD and the absence of medications specific to the latter (11, 12). However, several immunomodulatory drugs that are used in MS were associated with exacerbations in AQP4-ab positive NMOSD patients, which shifted the therapeutic approaches towards selecting medications based on known mechanisms of action for NMOSD and its pathophysiological characteristics (13). Consequently, MS drugs that mainly aim at B-cell depletion were adapted for use in NMOSD, e.g., mitoxantrone, rituximab, and azathioprine (14, 15).

Similarly, therapeutic approaches for MOGAD management have largely been extrapolated from NMOSD management approaches. Yet, emerging evidence indicates that there are distinct differences in how these disorders respond to various treatments (8). This highlights the need for a more tailored, condition-specific approach to managing MOGAD, rather than relying solely on the NMOSD treatment paradigm.

The worldwide estimated prevalence of NMOSD ranges from 0.5 to four per 100,000 and differs greatly among different races and regions (16). Regarding the Arab population, several hospital-based studies that aimed at quantification and characterization of NMOSD patients are published, with one being a retrospective multicenter study in The United Arab Emirates (UAE), which reported a prevalence of 1.76 per 100,000 for NMOSD (17). Additionally, a regional registry for NMOSD in the Arabian Gulf retrospectively reported on 144 NMOSD patients from five Arab countries, including The UAE, with the aim of describing demographical, clinical and radiological findings of NMOSD patients treated in 15 MS and neurology centers in the Arabian Gulf (18). Despite these efforts, data on NMOSD patients in UAE remain scarce. On the other hand, the worldwide prevalence of MOGAD is currently estimated at 1.3–2.5/100,000 (19); however, more accurate estimates are to be anticipated following the recent consensus on MOGAD’s diagnostic criteria. To date, there are no available epidemiological data on MOGAD in the UAE.

Consequently, and given the lack of a national registries for both demyelinating disorders (20), more data on the clinical presentation of NMOSD and MOGAD are needed to quantify their burden and establish standardized diagnosis and management strategies for them. Our study aimed to provide these data on a localized institutional level to precede and call for future generalized initiatives considering larger patient populations.

This was a retrospective, descriptive study on all patients having either NMOSD, based on the international NMOSD criteria (2), or MOGAD, identified from cohorts of previously considered seronegative NMOSD, treated and followed in Rashid hospital, a tertiary care center and MS research center in Dubai, UAE. Data collection was initiated on 30^th of January 2024, where patients’ data from the preceding five years were included.

Retrieved data comprised patients’ demographics, including age, gender, and ethnicity, as well as disease characteristics including age at onset, disease duration, and attacks presentations; treatments received and patients’ response to them; and laboratory and radiological findings. The collected data was then analyzed in the light of regional and international data.

Despite the rarity of CNS demyelination diseases, they still represent a great burden on both society and health care system; accordingly, our study discusses the demographics and patient characteristics observed for these diseases in a single UAE center. We identified 34 patients with clinical presentation fitting the clinical picture of non-multiple sclerosis atypical CNS inflammatory/demyelinating syndromes. Twenty-seven patients fulfilled the criteria of NMOSD, while seven patients tested positive for MOG-IgG antibodies fulfilling the criteria for an established MOGAD diagnosis. Regarding the NMOSD cohort, our results match the percentage for AQP4 seronegativity observed for NMOSD in literature, where five patients (19%) were AQP4-ab negative. Previous literature reported that about 20% to 30% of NMOSD patients have undetectable AQP4 antibodies (21). However, 34.7% (n = 50) of the 2019 Arabian Gulf NMOSD Registry population, and 53% (n = 50) of 103 NMOSD patients included in a population-based study in Iran, were AQP4-ab negative (18, 22). Consequently, and given that many NMOSD cases are still initially misdiagnosed as MS, applying the 2015 NMOSD diagnostic criteria to both AQP4 seropositive and seronegative patients allows for more efficient detection of NMOSD (2, 23).

On the other hand, in parallel with the evolvement of identification patterns for MOGAD, these patients are commonly retrospectively identified in NMOSD patient cohorts with AQP4-ab negative status (24). MOG-IgG antibodies were detected in 11.2% of clinically diagnosed NMOSD seronegative patients in a Chinese study of 125 NMOSD patients (25); similarly, a Brazilian study detected MOG-IgG antibodies in four of 47 patients with AQP4 seronegative NMOSD (26), while an Argentinean study reported that 27.7% of NMOSD seronegative patients tested positive for MOG-IgG antibodies (21). In the Arabian Gulf 2019 NMOSD registry, only 36% (n = 18) of the AQP4-seronegative patients were tested for MOG-IgG antibodies, with no positive results returned (18).

Two patients in our study cohort were double positive for both MOG-IgG and AQP4-IgG antibodies. This status was previously reported in an individual-case manner (4, 25, 27, 28); however, whether this population requires distinct management strategies is yet unknown. Similar to previous studies, the sample size in our study was too low to evaluate any of these outcomes in this population.

In our study, we compared patients with a definite NMOSD diagnosis to those with MOG antibodies—who were initially considered for an NMOSD diagnosis. For instance, the female predominance observed for NMOSD, particularly AQP4 seropositive cases, was not mirrored in the MOGAD cohort. Similar trend was observed in previous trials (8, 29, 30), where the female-to-male ratio can reach up to 3–9:1 in NMOSD, in contrary to a ratio of 1:1 in MOGAD (29, 31). Regarding the age of onset, our results go in line with literature suggesting that MOGAD usually present at a younger age than that for NMOSD (4). In contrast, in a study of 215 patients comparing NMOSD and MOGAD characteristics, the median onset age reported for MOGAD patients was 37.5 years, and that for APQ4 seropositive NMOSD was 37 years, while the lowest age of onset (32.5 years) was reported for the APQ4 seronegative study population (30). In a population-based study in Tehran, the mean age of onset was 31.5 years (range 8–61), which also matched that of the Arabian Gulf registry (31 ± 12 years) (18, 22).

In our cohort, the immunosuppressive maintenance therapies included azathioprine, IVMP, IVIG, mycophenolate mofetil (MMF), or rituximab as initial therapy, with a combination of the first and the latter in two NMOSD patients, and the combination of prednisolone and IVIG in another. In non-MS demyelinating disorders, treatment outcomes highly vary with treatment duration and disease onset and clinical picture. For instance, rituximab’s efficacy correlates with B cells levels, which are also subject to individual variations (32, 33). This calls for extensive understanding the relationships between patient variables and treatment outcomes; however, the small sample size in our study hindered the quantifications of such outcomes to drive the needed conclusions.

Regarding recovery degrees, reflected by the change in EDSS following the follow up period, MOGAD patients had greater median EDSS change and lower median EDSS scores at follow-up, which goes in line with previously reported data for MOGAD (34). For NMOSD, the seropositive group has demonstrated a slightly higher median EDSS change (0.8 vs. 0.5), which may be attributed to the lower baseline EDSS (2.8 vs. 4.0) (35). In literature, improvement in mean EDSS is comparable between seropositive and seronegative NMOSD populations (18).

In our study, most of the NMOSD AQP4 positive patients (82%) reported a positive response to the first line acute therapy IVMP; the same outcome was observed in the MOGAD cohort. Azathioprine was the most common maintenance therapy in both cohorts; however, our cohort of patients received different treatments with different durations, which hinders the interpretation of treatment outcomes in the light of other patient characteristics. This also may explain the variation observed in the number of relapses experienced by the study patients.

Our study has the limitation of small sample size, especially in the MOGAD cohort; therefore, care must be employed upon deriving definite conclusions where generalization is not optimal given the convenient sampling approach. However, our study lays the ground for further research to evaluate these outcomes in adequately powered comparative studies. Due to the retrospective nature of this study, correlation assessments for different clinical features, disease course, and the response to immunotherapies were not done. In that context, our study also highlights the limitations in logistical capabilities which complicated reporting on a larger array of disease features. Moreover, our study was conducted on an institutional level, while population-based studies are needed to assess the prevalence and incidence rates. However, conducting such studies for conditions like NMOSD and MOGAD may be challenging. This is because it can be difficult to establish consistent diagnostic criteria across different settings and to standardize the laboratory testing for the associated antibodies. Without these unifying factors, accurately assessing the true population-level epidemiology of NMOSD might not be feasible.

Our study provided valuable insights on the status of both NMOSD and MOGAD in our center; however, adequately powered prospective studies applying standardized definitions and methodologies are needed to further characterize NMOSD patients clinically and epidemiologically in the UAE.