A narrative review of identifying the culprit antibody in neuroimmune diseases: concept and clinical significance

Introduction

Autoantibodies are critical hallmarks of autoimmune diseases, which can be divided into pathogenic antibodies and concomitant antibodies according to their pathogenicity. Pathogenic antibodies reflect the etiology and pathogenesis of autoimmune diseases and are directly related to the onset, development, outcome, and therapeutic efficacy of the disease. Concomitant antibodies have no causal link with the disease, but can be used as an auxiliary indicator for disease diagnosis. Classical pathogenic antibodies in neuroimmune disease should meet the following criteria: (I) the antibodies exist in the serum or cerebrospinal fluid (CSF) of most patients or the titer of antibodies should be significantly higher in patients than that in healthy individuals; (II) the binding of the immunoglobulin to target antigen can be observed in the lesions by immunohistochemical staining; (III) removing circulating immunoglobulin by plasma exchange or absorption procedures can achieve therapeutic effect; (IV) the disease phenotype can be transferred to experimental animals by the injection of serum/CSF or purified immunoglobulin from patients and/or by sensitizing with corresponding target antigen in susceptible animals along with induction of the specific antibodies (1).

At present, neurologists rely much on the detection of autoantibodies for the diagnosis, severity assessment, prognosis, and efficacy evaluation of neuroimmune diseases. In recent years, the discovery of numerous antibodies dramatically promotes our understanding of neuroimmune diseases and improves disease diagnosis. At the same time, it brings confusion. Can every antibody be applied as an indicator for disease monitoring and therapeutic efficacy evaluation? Which antibody should be chosen for diagnosis or monitoring the treatment efficacy in the presence of two or more antibodies in the same patient? In this narrative review, we aimed to elucidate which antibody can be used as an indicator for a specific phenotype of neuroimmune diseases, including monitoring disease activity and efficacy assessment, and how to judge the causality with an integrative chain of evidence. The key clue of our discussion is the concept of “culprit antibody”. The importance of longitudinally dynamic identification of culprit antibodies is emphasized. We present the following article in accordance with the Narrative Review reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-21-1627/rc).

Methods

In this narrative review, we searched Medline, Embase, Web of Science, and CNKI from inception to February 2021 to acquire references regarding the relationship between phenotypes of neuroimmune diseases and antibodies. The search term “neuroimmunological/neuroimmune diseases and antibodies” were used. Titles and abstracts of resulting articles in English and Chinese language were screened and relevant articles on this research field were carefully scrutinized (Table S1).

Results

Till now, no reference on the culprit antibody was found. However, the references on the relationship between antibodies and phenotypes, disease development, disease monitoring, and treatment assessment were found in original research articles and case studies in neuroimmune diseases. Since this review is special in the nature of a proposal, the references are organized with the thread of culprit antibody.

The concept of culprit antibody, our proposal

Culprit antibody refers to the pathogenic antibodies that have causality with specific clinical phenotype, similar to “culprit vessels” applied in ischemic stroke or “criminal vessels” in ischemic heart disease. Although atherosclerosis can be present in major arteries throughout our body and may potentially progress into symptomatic, not all arteries with atherosclerosis will cause symptoms and injuries to the target organs. At the same time, other mechanisms (embolism or vasospasm) may lead to the same or similar symptoms. In some patients, various mechanisms coexist. Pathogenic antibodies may also be latent during the early course of neuroimmune diseases. Moreover, there is ample evidence that multiple antibodies can be found coexisting in patients with a specific neuroimmune phenotype, and various neuroimmune diseases may coexist in one patient. It is important to catch the exact originator.

There are some points for this concept:

• The culprit antibody should be a pathogenic antibody instead of a concomitant antibody, its titer correlates with the disease severity and treatment response, and the change of its titers is closely related to the prognosis. For example, anti-acetylcholine receptor (AChR) antibodies and anti-muscle specific kinase (MuSK) antibodies are considered pathogenic autoantibodies in myasthenia gravis (MG), which affect the transmission of neuromuscular junctions, leading to the fatigue symptoms of MG. Their titers are correlated with the severity of MG in a given MG patient during the long-term follow-up (2). However, although the disease of MG patients with anti-titin antibodies may be more severe than those without this antibody, the underlying causal relationship is that the severity is associated with thymoma, which constitutes a more complex and severe immune dysfunction and is also closely associated with the titin antibody. No substantial evidence supports that the presence of titin antibodies directly leads to MG symptoms (3). The titin antibody does not fluctuate in parallel with the severity of MG, so it cannot be used as an indicator of disease severity and prognosis of MG (4). Another example is the anti-N-methyl-D-aspartate receptor (NMDAR) antibody, which can block and internalize NMDAR, leading to neuropsychiatric symptoms, memory loss, seizures, etc., and is considered a pathogenic autoantibody involving in NMDAR encephalitis. The titer of NMDAR antibody in CSF is closely related to the disease course and prognosis of NMDAR encephalitis patients (5-7). However, there is a frequent concomitant presence of low-titer anti-myelin oligodendrocyte glycoprotein (MOG) antibodies in NMDAR encephalitis, which may disappear in short term and is not directly correlated with the severity and prognosis of encephalitis.
• A pathogenic antibody can be identified as a culprit antibody only when its causal relationship with a specific clinical phenotype is confirmed. Again, the anti-MOG antibody is an example. It can be detected transitorily in some patients with only NMDAR encephalitis phenotype and without any clinical and radiological signs of inflammatory demyelination of the optic nerve, brain, and spinal cord (8). A secondary immune response against the debris from injuries of encephalitis might be its origin.
• As the basis of culprit antibody, a known “evidence chain” helps us to understand why it is the culprit, including antigen recognition, antibody production, antigen-antibody binding and subsequent injuries or impairment of the structure and function of the nervous system, and specific correlation with clinical phenotypes. The role of the “evidence chain” can be traced back to the historical research of anti-NMDAR encephalitis. In 1968, Corsellis et al. proposed the term limbic encephalitis (LE) (9). After that, Vitaliani et al. reported autoimmune encephalitis with teratoma and proposed the term teratoma-associated encephalitis (10). Finally, Dalmau et al. determined the NR1/NR2 domain in NMDAR protein on the surface of the hippocampal neuron membrane was the main antigenic epitope, and serum from these patients specifically cross-react with both teratoma and brain neurons and proposed the disease of anti-NMDAR encephalitis (11). The purified human NMDAR antibody can also induce phenotypes similar to the clinical manifestations of human disease in animal models, which provides evidence for antibody pathogenicity (12). These provide the basis for the identification that the NMDAR antibody is the culprit antibody against NMDAR encephalitis.

The link between culprit antibody and clinical phenotypes

It is known that the production of pathogenic antibodies against antigens in the nervous system can be provoked by infection, tumor, and other predisposing factors, which leads to the exposure of antigen epitopes in normal nerve tissues (such as in post-infectious NADAR encephalitis) or ectopic expression of cross-reactive antigens in other tissues (such as in teratoma-associated NMDAR encephalitis), or as a systematic immune response against a premorbid respiratory/gastrointestinal infection [such as in post/para-infectious Guillain-Barre syndrome (GBS)] or an exogenetic substance (such as in GBS after ganglioside treatment). These sensitized immune cells are carried to local lymph nodes, where B cells differentiate into plasma cells and memory B cells via the interaction with antigen and CD4⁺ T cells. Plasma cells produce antibodies. Autoantibodies can directly act on the peripheral nervous system or neuromuscular junction with blood circulation to produce clinical symptoms, as anti-ganglioside antibodies or anti-AChR antibodies in GBS or MG (4,13); they can also take effect on the central nervous system (CNS) through the damaged blood-brain barrier, as anti-aquaporin-4 (AQP-4) antibodies in neuromyelitis optica spectrum disorder (NMOSD) (14,15). Some memory B cells can enter the brain and form plasma cells through re-stimulation and antigen, driving to produce antibodies, such as in the case of anti-NMDAR encephalitis (16). Although pathogenic autoantibodies have the potential to induce clinical symptoms, whether they lead to symptoms is also related to antibody titer, the duration of their presence, their capacity to reach the target tissue, and the ability of the target tissue to compensate for damage.

The clinical syndrome of an individual patient at a certain stage of a disease may be caused by the binding of the “culprit antibody” to the corresponding target tissue and subsequent tissue damage or impairment of its function. The determination of a “culprit antibody” depends on the localization diagnosis procedure. The antibodies bind to various specific locations within the nervous system, leading to the relevant neurological syndromes and signs, which constitute their phenotypes. For example, the binding of AQP-4 antibody the area postrema causes demyelination, which presents as the area postrema syndrome characterized by intractable hiccup or vomiting (17,18); the binding of anti-sulfatide antibodies to unmyelinated nerve fibers in peripheral nerves which conduct the pain sensation can lead to painful peripheral neuropathy (19), and the binding of leucine-rich glioma inactivated gene-1 (LGI-1) antibody to its ligand in the brain area which modulates the voluntary movement can lead to encephalitis characterized by faciobrachial dystonic seizure (FBDS) (20).

Culprit antibody and core phenotype

In the phenotypic study of neuroimmune diseases, specific phenotypes are identified according to the common characteristics of a specific disease, that is, a constellation of the most common symptoms and abnormal auxiliary investigations reflecting the underlying pathological process, which constitute the “prototype” of the disease. With the screening for disease-related antibodies, specific disease-associated antibodies are discovered in patients with typical phenotypes. Then through the direct evidence of the relationship between antibody level and symptom severity, and the binding of the antibody to nerve tissue, the pathogenic antibody is finally identified as the causal factor because it determines the relevant phenotype. For example, anti-GQ1b antibody is detected in Miller Fisher syndrome (MFS) characterized by extraocular palsy, decreased tendon reflex, and ataxia (21), and anti-AQP-4 antibody was detected in patients with Devic’s disease characterized by involvement of optic nerve and longitudinal extensive lesions of the spinal cord (22). Furthermore, using these pathogenic antibodies as clues, “variant” or “atypical clinical manifestations” are found in patients whose disease course and treatment efficacy are consistent with autoimmune diseases and whose corresponding antigens are expressed in the affected locations which are related to clinical symptoms, clarifying the causal relationship between the atypical phenotypes and these antibodies. Finally, based on the research on the antigen location and antibody accessibility, the screening of these antibodies in patients with atypical phenotypes obtained a broader boundary of the clinical phenotype of the prototypic disease and formed new sub-phenotypes when enough patients were reported. For example, GQ1b antibody links MFS, Bickerstaff brainstem encephalitis, and GBS to constitute GQ1b antibody syndrome within the spectrum of GBS (23,24), and AQP-4 antibody links the area postrema syndrome, extensive white matter lesions in the brain, recurrent optic neuritis and recurrent myelitis to constitute NMOSD (22).

Overlapping sub-phenotypes exist within a disease entity. As is illustrated in Figure 1, four different sub-phenotypes of GBS, demyelinating GBS, axonal GBS, MFS and pharyngeal-cervical-bronchial GBS variant (PCB) constitute a phenotypic spectrum of GBS, there are distinct typical sub-phenotypes and overlap phenotypes. Typical sub-phenotypes including MFS, axonal GBS, and PCB, have their relatively specific pathogenic antibodies, GQ1b/GD1b antibodies, GM1 antibodies, and GT1a antibodies, although there is some cross-reactivity between some antibodies (e.g., the fine specificity of GQ1b and GT1a antibodies, there is cross-reactivity in some patients while no cross-reactivity in others) (25). Under these circumstances, it is essential to determine the core sub-phenotype within a disease entity and the relevant culprit antibody, for disease monitoring and prognosis. When MFS is the predominant phenotype, a low-titer GM1 antibody is considered a concomitant antibody. However, if there is a high titer GM1 antibody, care should be taken to search for potential axonal involvement, including the rapidly recovered axonal type of reversible conduction block (26). Otherwise, the accuracy of GM1 antibody testing should be questioned.

Figure 1 An example for explaining “core sub-phenotypes” and “culprit antibody”. This figure illustrates four sub-phenotypes- demyelinating GBS, axonal GBS, MFS and PCB, which may overlap both in phenotypic features and in culprit antibodies. Although anti-GQ1b antibody is always associated with MFS characterized by extraocular palsy, decreased tendon reflex and ataxia, it is also detected in the overlapping GBS-MFS phenotype and in the overlapping MFS-PCB phenotype. There are still other overlapping phenotypes composed of these core sub-phenotypes. GBS, Guillain-Barre syndrome; MFS, Miller Fisher syndrome; PCB, pharyngeal-cervical-bronchial variant.

Thus, the pathogenic antibodies discovered from the prototype disease expand the boundaries of the disease entity and form new sub-phenotypes. These sub-phenotypes are not the same as the prototype phenotype, but there are some overlaps between them. The key point is again that the antigen-antibody binding determines the involvement location in the nervous system, and the pathogenic antibody against the target epitope leads to the corresponding clinical manifestations. When there are enough data on typical sub-phenotypes which have a specific constellation of clinical features, several “core phenotypes” are formed under a disease spectrum (27).

Culprit antibody and antibody overlapping syndrome

The culprit antibody is particularly important in diagnosing and treating patients with two or more antibodies, which is called antibody overlapping syndrome. Antibody overlapping syndrome of the nervous system is not uncommon in clinical practice. A single-center study found that 63/193 (33%) thymoma patients had two or more antibodies, including nicotinic AChR antibody, ganglionic AChR antibody, LGI1 antibody, Caspr2 antibody, α-amino-3-hydroxy-5-methyl-4-isoxazole-pro pionic acid receptor (AMPAR) antibody, AQP-4 antibody, γ⁃aminobutyric acid type A receptor (GABA_AR) antibody, and glycine receptor (GlyR) antibody. Among them, 26 patients showed typical MG phenotypes overlapped with phenotypes of other neuroimmune diseases (28). The overlap of anti-NMDAR encephalitis and demyelinating diseases of the CNS is also common (8,29). In one study of Huashan Hospital, among 87 cases of anti-MOG antibody-associated demyelinating diseases, 18 patients had clinical manifestations of encephalitis, 16 of which met the diagnosis criteria of autoimmune encephalitis, and five patients were positive for NMDAR antibody (29). A comprehensive summary of reported cases revealed that 3.3% (23/691) of patients with NMDAR encephalitis had typical imaging manifestations or clinical symptoms of CNS demyelination, including 9 cases of positive AQP-4 antibody and 9 cases of positive MOG antibody. CNS demyelination occurred simultaneously with (11/691), before or after (12/691) NMDAR encephalitis. In these NMDAR encephalitis patients, only one was found with teratoma, suggesting that non-teratoma etiology leads to both encephalitis and demyelination (30). By using cluster and factorial analysis, we found that the NMDAR antibody, the AQP-4 or MOG antibody has its specific core phenotypes in patients with antibody overlapping syndrome, which was still consistent with the phenotypes related to specific antibodies (31). This study supports the concept of culprit antibodies in real-world practice.

Coexistence of multiple culprit antibodies at different disease stages

During the course of an antibody overlapping syndrome of the nervous system, different antibodies may appear successively, and the clinical phenotypes at different disease stages are associated with the corresponding culprit antibody. We reported a patient who was admitted with apathy, behavioral abnormalities, and transient inarticulate speech. PET-CT showed a thymoma. Both cell-based assay (CBA) and tissue-based assay (TBA) were positive for serum AMPAR antibody. The patient was diagnosed with autoimmune encephalitis and discharged after clinical improvement. No AChR antibody was found in the serum at that time. Nine months later, the patient was hospitalized due to dysphagia and slurring speech after fatigue, with no overt clinical symptoms of autoimmune encephalitis. A diagnosis of MG was made. At this time, besides positive AMPAR antibody, serum AChR antibody and titin antibody were also found positive (32). AMPAR antibody and AChR antibody play their leading role in different stages of this antibody overlapping syndrome. A retrospective study of 16 patients overlapping with MG and NMOSD found that serum AQP-4 antibodies can be detected positive 4 to 16 years before the onset of NMOSD. In five patients with MG prior to NMOSD, the serum AChR antibody titer was the highest at the onset of MG and gradually decreased with treatment. Meanwhile, the serum AQP-4 antibody titer showed a gradual upward trend, peaking at the first onset of NMOSD (33). AQP-4 antibody and AChR antibody are likely to be detected both positive at a particular stage of the disease course in such patients. Hence, different culprit antibodies predominated in different stages of the same patients. Since the culprit antibody is parallel to the disease severity, the determination of the culprit antibody will not only help the diagnosis but also have a role in monitoring disease activity, evaluating treatment efficacy, and implying the prognosis.

Conclusions

It is undeniable that the speed of discovering autoantibodies of neuroimmune diseases is unprecedented. At present, the confirmation of the pathogenicity of these new antibodies often employs arduous and prospective research instead of retrospective analysis, which embodies the essence of translational medicine—from bed to bench and then from the bench back to bed. This constitutes closed-loop clinical research from antibody screening, antibody subtype identification, clinical phenotype characterization, and antigenic epitope location with animal models and cell model verification. Identifying new pathogenic antibodies and rigorously verifying their potential significance as culprit antibodies will be a long-term hotspot and focus in the field of neuroimmunology in the future.

With the application of massive antibody testing, more and more cases with multiple antibodies will be found in patients who are diagnosed with one neuroimmune disease initially. In patients with neuroimmune antibody overlapping syndrome, it is necessary to make independent diagnoses of various neuroimmune diseases according to their relevant diagnostic criteria. Moreover, it is essential to evaluate the priority, severity, and urgency of various overlapping phenotypes in clinical practice. The concept of culprit antibodies will help to select the optimal antibodies for monitoring disease activity precisely and avoid wasting the cost on repeated testing of the unrelated antibodies.

The narrative review provides a preliminary outline of the concept, the causal relationship with clinical phenotypes, and the significance of the culprit antibody. Although the concept could not explain all scenarios in neuroimmune diseases, we believe the concept will promote further investigation and clinical application of newly-found antibodies.

Acknowledgments

We thank Wei Qiu (Department of Neurology, The Third Affiliated Hospital of Sun Yat-Sen University), Yan Xu (Department of Neurology, Peking Union Medical College Hospital), and Sheng Chen (Department of Neurology, Shanghai Ruijin Hospital, Shanghai Jiao Tong University School of Medicine) for their active discussion and advice on this manuscript.

Funding: This study was supported by the Clinical Research Plan of SHDC (No. SHDC2020CR2027B to X Chen), National and Provincial Multi-Disciplinary Cooperation in Diagnosis and Treatment of Major Diseases Capacity Improvement Project (Shanghai Municipal Health Commission), and the National Natural Science Foundation of China (No. 82171397 to HF Li).

Footnote

Provenance and Peer Review: This article was commissioned by the editorial office, Annals of Translational Medicine for the series “Laboratory Investigations in Neuroimmunological Diseases and Their Clinical Significance”. The article has undergone external peer review.

Reporting Checklist: The authors have completed the Narrative Review reporting checklist. Available at https://atm.amegroups.com/article/view/10.21037/atm-21-1627/rc

Conflicts of Interest: Both authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-21-1627/coif). The series “Laboratory Investigations in Neuroimmunological Diseases and Their Clinical Significance” was commissioned by the editorial office without any funding or sponsorship. XC and HFL served as the unpaid Guest Editors of the series. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Archelos JJ, Hartung HP. Pathogenetic role of autoantibodies in neurological diseases. Trends Neurosci 2000;23:317-27. [Crossref] [PubMed]
2. Gilhus NE, Skeie GO, Romi F, et al. Myasthenia gravis - autoantibody characteristics and their implications for therapy. Nat Rev Neurol 2016;12:259-68. [Crossref] [PubMed]
3. Chen XJ, Qiao J, Xiao BG, et al. The significance of titin antibodies in myasthenia gravis--correlation with thymoma and severity of myasthenia gravis. J Neurol 2004;251:1006-11. [Crossref] [PubMed]
4. Lazaridis K, Tzartos SJ. Myasthenia Gravis: Autoantibody Specificities and Their Role in MG Management. Front Neurol 2020;11:596981. [Crossref] [PubMed]
5. Frechette ES, Zhou L, Galetta SL, et al. Prolonged follow-up and CSF antibody titers in a patient with anti-NMDA receptor encephalitis. Neurology 2011;76:S64-6. [Crossref] [PubMed]
6. Gresa-Arribas N, Titulaer MJ, Torrents A, et al. Antibody titres at diagnosis and during follow-up of anti-NMDA receptor encephalitis: a retrospective study. Lancet Neurol 2014;13:167-77. [Crossref] [PubMed]
7. Brooks J, Yarbrough ML, Bucelli RC, et al. Testing for N-methyl-d-aspartate Receptor Autoantibodies in Clinical Practice. Can J Neurol Sci 2020;47:69-76. [Crossref] [PubMed]
8. Martinez-Hernandez E, Guasp M, García-Serra A, et al. Clinical significance of anti-NMDAR concurrent with glial or neuronal surface antibodies. Neurology 2020;94:e2302-10. [Crossref] [PubMed]
9. Corsellis JA, Goldberg GJ, Norton AR. "Limbic encephalitis" and its association with carcinoma. Brain 1968;91:481-96. [Crossref] [PubMed]
10. Vitaliani R, Mason W, Ances B, et al. Paraneoplastic encephalitis, psychiatric symptoms, and hypoventilation in ovarian teratoma. Ann Neurol 2005;58:594-604. [Crossref] [PubMed]
11. Dalmau J, Tüzün E, Wu HY, et al. Paraneoplastic anti-N-methyl-D-aspartate receptor encephalitis associated with ovarian teratoma. Ann Neurol 2007;61:25-36. [Crossref] [PubMed]
12. Giannoccaro MP, Wright SK, Vincent A. In vivo Mechanisms of Antibody-Mediated Neurological Disorders: Animal Models and Potential Implications. Front Neurol 2019;10:1394. [Crossref] [PubMed]
13. Willison HJ, Jacobs BC, van Doorn PA. Guillain-Barré syndrome. Lancet 2016;388:717-27. [Crossref] [PubMed]
14. Shimizu F, Schaller KL, Owens GP, et al. Glucose-regulated protein 78 autoantibody associates with blood-brain barrier disruption in neuromyelitis optica. Sci Transl Med 2017;9:eaai9111. [Crossref] [PubMed]
15. Wu Y, Zhong L, Geng J. Neuromyelitis optica spectrum disorder: Pathogenesis, treatment, and experimental models. Mult Scler Relat Disord 2019;27:412-8. [Crossref] [PubMed]
16. Huang YQ, Xiong H. Anti-NMDA receptor encephalitis: a review of mechanistic studies. Int J Physiol Pathophysiol Pharmacol 2021;13:1-11. [PubMed]
17. Dubey D, Pittock SJ, Krecke KN, et al. Association of Extension of Cervical Cord Lesion and Area Postrema Syndrome With Neuromyelitis Optica Spectrum Disorder. JAMA Neurol 2017;74:359-61. [Crossref] [PubMed]
18. Ciron J, Sourdrille F, Biotti D, et al. Area postrema syndrome: Another feature of anti-GFAP encephalomyelitis. Mult Scler 2020;26:253-5. [Crossref] [PubMed]
19. Dabby R, Weimer LH, Hays AP, et al. Antisulfatide antibodies in neuropathy: clinical and electrophysiologic correlates. Neurology 2000;54:1448-52. [Crossref] [PubMed]
20. Binks SNM, Klein CJ, Waters P, et al. LGI1, CASPR2 and related antibodies: a molecular evolution of the phenotypes. J Neurol Neurosurg Psychiatry 2018;89:526-34. [Crossref] [PubMed]
21. Chiba A, Kusunoki S, Shimizu T, et al. Serum IgG antibody to ganglioside GQ1b is a possible marker of Miller Fisher syndrome. Ann Neurol 1992;31:677-9. [Crossref] [PubMed]
22. Kim SH, Kim W, Li XF, et al. Clinical spectrum of CNS aquaporin-4 autoimmunity. Neurology 2012;78:1179-85. [Crossref] [PubMed]
23. Lee SH, Lim GH, Kim JS, et al. Acute ophthalmoplegia (without ataxia) associated with anti-GQ1b antibody. Neurology 2008;71:426-9. [Crossref] [PubMed]
24. de Bruyn A, Poesen K, Bossuyt X, et al. Clinical spectrum of the anti-GQ1b antibody syndrome: a case series of eight patients. Acta Neurol Belg 2019;119:29-36. [Crossref] [PubMed]
25. Nagashima T, Koga M, Odaka M, et al. Clinical correlates of serum anti-GT1a IgG antibodies. J Neurol Sci 2004;219:139-45. [Crossref] [PubMed]
26. Umapathi T, Tan EY, Kokubun N, et al. Non-demyelinating, reversible conduction failure in Fisher syndrome and related disorders. J Neurol Neurosurg Psychiatry 2012;83:941-8. [Crossref] [PubMed]
27. Li H. A preliminary discussion on phenotypes of neuroimmunological diseases. Journal of Neurology and Neurorehabilitation 2019;15:22-31.
28. Zekeridou A, McKeon A, Lennon VA. Frequency of Synaptic Autoantibody Accompaniments and Neurological Manifestations of Thymoma. JAMA Neurol 2016;73:853-9. [Crossref] [PubMed]
29. Wang L. Encephalitis is an important clinical component of myelin oligodendrocyte glycoprotein antibody associated demyelination: a single-center cohort study in Shanghai, China. Eur J Neurol 2019;26:168-74. [Crossref] [PubMed]
30. Huang Y, Wang Q, Zeng S, et al. Case Report: Overlapping Multiple Sclerosis With Anti-N-Methyl-D-Aspartate Receptor Encephalitis: A Case Report and Review of Literature. Front Immunol 2020;11:595417. [Crossref] [PubMed]
31. Shen MH, Yu H, Liu XN, et al. Analysis of clinical characteristics of patients with positive anti-myelin oligodendrocyte glycoprotein antibody and anti-N-methyl-D-aspartate receptor antibody. Chinese Journal of Neurology 2021;54:898-907.
32. Li X, Mao YT, Wu JJ, et al. Anti-AMPA receptor encephalitis associated with thymomatous myasthenia gravis. J Neuroimmunol 2015;281:35-7. [Crossref] [PubMed]
33. Leite MI, Coutinho E, Lana-Peixoto M, et al. Myasthenia gravis and neuromyelitis optica spectrum disorder: a multicenter study of 16 patients. Neurology 2012;78:1601-7. [Crossref] [PubMed]