4 Discussion
The dual positivity of anti-MOG and anti-NMDAR overlapping syndrome is an increasingly recognized but rare disease entity that has been reported in only a few cases [12, 14, 15, 17, 18]. In a cohort study by Fan et al., 5 of 42 (11.9%) patients with MOG-antibody disease were identified as NMDAR encephalitis, the frequency of which was higher than that of NMOSD overlapping with NMDAR encephalitis (0.6%) [11]. However, the number of cases is limited and mainly restricted to children. Furthermore, a large-scale study investigating the co-existence of NMDAR antibodies in 376 MOG-IgG1 patients concluded that testing for MOG-IgG1 and NMDAR-IgG is essential in patients with encephalopathy and demyelinating syndromes [19]. Thus, we realized that the rate of coexisting MOGAD and anti-NMDAR encephalitis was underestimated because of the complex symptoms and heterogeneity of clinical manifestations [12].
Unlike the typical presentation of MOGAD as ON and TM in adults and ADEM in children, common clinical manifestations of the overlapping syndrome include headache, fever, diplopia, vision loss, palsy, aphasia, psychosis, seizures, and confusion. We observed a combination of focal neurological deficits and encephalopathy symptoms in these patients, which may provide clues for recognition of MOGAD overlapped NMDAR encephalitis. As visual impairment and myelopathy are rare in patients with NMDAR encephalitis [5], epilepsy and behavioral disturbance are less frequent in MOGAD [4]; therefore, we suggest that patients suffered from encephalitis and demyelination testing for both NMDAR and MOG antibodies.
Typical MRI findings in overlapping syndromes are differentiated from MOGAD with lesions involved anterior optic nerve bilaterally [2], or from NMDAR encephalitis with lesions involved temporal and hippocampal regions asymmetrically [5]. Brainstem was the most common involvement in our study, followed by lesions in the cortex and deep gray matter. Furthermore, a few patients showed both supratentorial and infratentorial lesions, which may be warning signs of MOGAD overlapping with anti-NMDAR encephalitis. A case report of a psychiatric patient with triple antibodies against anti-NMDAR, anti-CASPR2, and MOG mentioned that bilateral cingulate and hippocampal lesions might be imaging clues suggestive of coexisting antibodies [20]. However, in our study, the involvement of the cingulate gyrus was observed in only one case and that of the hippocampus in two cases, indicating that cingulate gyrus involvement occurred less frequently and could occur unilaterally.
MOGAD can present as either a monophasic or relapsing disease [3]. However, in our cohort, 100% of the patients developed relapse or multiple relapse attacks, indicating a high recurrence risk of the overlapping syndrome. By analyzing the clinical course of each case, we found that MOG positivity occurred before (10%), simultaneously (40%), or after (50%) NMDAR encephalitis, which may indicate pathophysiological differences. Viral infection is considered a precipitating factor for AE [21]. Mariotto et al. reported that 45% of patients with MOGAD had prodromal symptoms or previous infectious processes [22]. In our study, the inducing factors included influenza-like symptoms in one case (pat 1) and viral infections (Herpes S Virus and Epstein–Barr virus) with evidence from CSF next-generation sequencing in three cases (pat 5–7). Brain MRI of Patient 4 showed severe blood-brain barrier (BBB) damage with marked meningeal enhancement after herpes simplex virus infection of the CNS. In addition, our study found that other causes of disease relapse may point to tapering or cessation of steroids, as 60% of patients experienced relapses due to inappropriate glucocorticoid adjustments. When analyzing antibody dynamics in patients, we found that clinical deterioration was associated with increased titers of the responsible antibody. And between these two antibodies, MOG-IgG was more pathogenic than NMDAR-IgG.
The pathogenesis of overlapping autoimmune syndromes remains poorly understood. Based on the observations from our cases and previous reports, we hypothetically propose the possible pathogenesis of MOGAD overlapped with anti-NMADR encephalitis. First, MOG is specifically located on the outer surface of oligodendrocytes in the CNS [23, 24], and NMDAR is expressed on neurons, oligodendrocytes, astrocytes, and excitatory glutamate synapses [25]. This is the biological structural basis of the occurrence of immune cross-reactivity and explains the dual antibody positivity simultaneously in the first episode. Second, viruses can trigger CNS inflammation and immune dysfunction, resulting in NMDAR encephalitis [21]. An excessive inflammatory response leads to BBB breakdown, followed by MOG antigen leakage into the peripheral blood and serum-positive MOG antibody generation. This mechanism reveals an NMDAR-positive course accompanied by or followed by MOGAD, which is the major phenotype of overlapping syndromes. Third, multiple cases showed that the immune system could rebuild and renovate itself when the dosage of immunotherapy is decreased. Meanwhile, the antigenicity of the self-structure could increase, giving rise to immune disorder [26-28]. These theories of immune restitution may account for relapse attacks during steroid tapering or cessation.
In accordance with previous studies [15, 29], patients with overlapping syndromes had good responses to first-line treatments, such as IVMP, IVIG, PE, and IA, with improved mRS scores and adsorbed lesions on MRI. However, if glucocorticoids are reduced too quickly or discontinued, relapse may occur [15]. However, the optimal timing of immunosuppressant administration remains unclear. In our study, 90% patients were given azathioprine for long-term therapy at their first episode. Then they experienced disease relapse at 6.7 months as median interval. It means the relapse might occur when glucocorticoids tapering, in the meanwhile, immunosuppressants unable to take effects sufficiently. Hence, we recommend adding second-line treatments once diagnosed as MOGAD overlapped with anti-NMADR encephalitis. Moreover, high-efficiency immunosuppressants (rituximab or mycophenolate mofetil) were suggested to reduce further relapse risks. In general, patients recovered well after appropriate first-line treatment with immunosuppressants [30, 31], and our patients showed no significant disability at the final follow-up (average time: 25.4 months). Additionally, we provided lymphocyte subset data for these patients, which is the first report of MOGAD overlapping with anti-NMDAR encephalitis. By comparing the percentage of T cells subtype before and after immunosuppressant treatment, we found that CD3+ and CD4+ T cell counts increased after therapy. However, detailed information on T and B lymphocyte subsets requires further investigation.
There are some limitations in the present study. First, the sample size was limited owing to the low incidence of the disease, although we enrolled all patients diagnosed within the last five years. Second, the retrospective nature of this study determined that some treatments were uncontrollable, which might have caused bias. Third, in real-world clinical practice, data on lymphocyte subsets were not perfect. Only a subset of patients underwent lymphocyte subsets test, and we could not obtain sufficient data to conduct a statistical analysis of T lymphocyte subgroups between different immunosuppressive drugs.
In conclusion, the main clinical feature of MOGAD overlapping with NMDAR encephalitis is encephalopathy symptoms associated with focal neurological deficits. Brainstem lesions combined with cortical involvements on MRI may be warning signs of this overlapping syndrome. Because the relapse rate is high, we believe that early diagnosis and timely treatment with high-efficiency immunosuppressants at onset would be beneficial for these patients.