Neuromyelitis optica spectrum disorder (NMOSD), an autoimmune inflammatory disease, is commonly seen in Asian populations. The number of patients in China is currently approximately 100,000 (1, 2). Studies have demonstrated that 90% of NMOSD patients show a remission-recurrence course (3). Notably, neurological disability can occur after one attack and accumulate with each relapse in patients with NMOSD. However, the evolving course and treatment responses in patients with acute episodes of NMOSD remain unclear. Based on a small case series and experience in other immunological diseases, intravenous methylprednisolone pulse therapy (IVMP) is currently recommended as a first-line regimen for acute onset of NMOSD (4–7). However, IVMP is effective in only 60%-80% of NMOSD patients (8, 9). For patients with a poor response to IVMP treatment, plasma exchange (PE) or a large dose of intravenous immunoglobulin (IVIg) may be effective for NMOSD attacks (10, 11). In general, there is still a lack of data on the therapeutic options for NMOSD attacks and the corresponding efficacy.

Current investigations have advocated neuroinflammatory changes in NMOSD. However, there is a paucity of data on the changes in serum cytokines and chemokines in NMOSD patients (12–15). Based on the available study, we propose that cytokines related to the activation of macrophages, effectors of T and B lymphocytes and chemokines are involved in the mechanism of NMOSD. As the neurofilament light chain (NfL) shows a potential role in the pathogenesis of NMOSD (16), we measured serum levels of these cytokines/chemokines as well as NfL to assess their usefulness as biomarkers in clinical practice.

Therefore, this prospective observational cohort trial was conducted to examine treatment responses and outcomes with new-onset and relapsing NMOSD. We also explored possible serum biomarkers of disease severity and prognosis.

This is a prospective study that recruited NMOSD patients in the acute phase to observe the characteristics and outcomes of attacks. In terms of subtype classification, MY accounted for the most cases, followed by ON and brain/brainstem syndromes, which is in line with previous studies (8, 23). Regarding therapeutic options, a majority of patients (71%) were treated with IVMP alone, and only 8% opted for PE in this trial, exhibiting a notable difference from other studies. Apheresis therapies such as PE and immune adsorption account for 20% or more of acute treatments in other studies (8, 24). The reasons are complicated and are mainly due to the relatively mild severity of the patients in our cohort.

In the whole cohort, the EDSS scores continuously and significantly decreased until the 3-month visit (3.0 [2.0–4.0]) compared with the onset scores (4.0 [3.0–5.5]). Since EDSS is insensitive to vision change, we use logMAR to evaluate the improvement of vision defect (25). The visual improvement reached a plateau at 1 month and could not be further improved until the 6-month point, which was partially consistent with the EDSS scores. We then explored the remission rate in all patients. Thirty-nine percent (38%) of the subjects showed remission at discharge, and 63.3% showed remission at 1 month after acute remedy. Overall, the outcome of attacks was generally moderate. This is fractionally different from other cohorts with NMOSD attacks. A total of 21.6% of NMO attacks showed a full recovery, and 6% showed no improvement at all after the acute treatment course (8). Suitable outcome criteria designed for clinical symptoms in NMOSD attacks are scarce, and the formula we applied in this study based on the degree of disability (20). Hence, the variability of outcome between this and other studies may be attributed to the differences in the evaluation criteria and the condition of the enrolled population.

We observed the efficacy measured by remission rate across phenotypes. Attacks such as ON, MY, and ON combined with MY and other classifications had the same remission rate at discharge. Of note, isolated ON was associated with particularly unfavorable outcomes and attacks involving MY had a higher likelihood of achieving remission at the 1-month follow-up. This might indicate that the recovery of vision impairment is slower than that of paralysis, sensory loss and bladder dysfunction. Differently, manifestation as MY or bilateral ON was associated with particularly unfavorable outcomes after the course of therapies among NMO patients in Germany (8). We also compared the efficacy of different treatment options, but no significant difference was found. Although studies ascertained the importance of PE as a therapeutic intervention in acute attacks (10, 26), PE was not associated with an additional curative effect when compared to IVMP alone in our trial. These might be due to the limited number of cases in some treatment options and/or possible baseline differences between distinct subgroups, an inherited disadvantage of an observational study design. In fact, PE is recommended as soon as possible for patients with severe attacks of CNS demyelination, with or without a combination of IVMP (27, 28).

The present study provides data on a second clinical event following acute NMOSD attacks in a relatively short period. In detail, approximately 9% of patients relapsed within 3 months, and 13% relapsed within 6 months during this trial. And the previous annualized relapse rate of this cohort was 0.8(IQR 0.4–1.2) per year, which seemed higher than that in follow-up phase after this episode. And we might ascribe this phenomenon to the better compliance of the patients after enrollment. From a long-term perspective, approximately 60% of patients relapse within 1 year, while 90% relapse within 3 years in another study (3). Scholars also reported that 86.6% of pediatric NMOSD patients had a second clinical event after a follow-up of 2–10 years, with a median disease duration of 4 years (29). Those patients who experienced recurrence in our trial had no special attributes and were not clustered in onset clinical subtype, AQP4 status or treatment modality. A recent study suggested that approximately half of relapses occurred within 12 months and presented with similar manifestations as the last clinical attack (23).

B-cell modulation, imbalance of Th1/Th2 and upregulation of Th17 are essential factors for developing NMO inflammatory lesions (30, 31). Cytokines, including IL-4, IL-10, IL-13 and IL-Ra, are Th2-type cytokines and are beneficial in CNS inflammatory disorders (13, 32). Our study reveals that a high level of Th2-type cytokines (PC1) composed of these four cytokines rather than other components predicts remission at the 1-month visit but not at discharge. This suggests a protective role of a Th2 dominant response in the cascade of immune events involved in NMOSD pathogenesis. Moreover, the prognostic effect was not achieved at the end of acute management, which might be attributed to the delayed effect of Th2 cells in the natural course of NMOSD attack. Interestingly, there was an increased peripheral Th2 proportion (CD4⁺IL-4⁺) in the attack phase compared with the remission phase, and the Th2 proportion was associated with multiple lesion locations in AQP4-IgG-positive patients (33). These results both supported a potential role of Th2 cells in the pathogenesis and progression of NMOSD. More studies are required to further elucidate the immunological characteristics of NMOSD. Our study supports that the Th2 cytokine profile, including IL-4, IL-10, IL-13 and IL-Ra, might improve our ability to monitor the treatment response in acute NMOSD episodes.

NfL is a structural element of neurons that results from neuroaxonal damage and appears in CSF and blood in neurological disease (34–37). In our trial, sNfL levels were positively associated with EDSS scores but irrelevant to the degree of recovery from NMOSD relapses. In addition, we demonstrated that sNfL levels could not predict subsequent relapses within 6 months in patients with NMOSD. It was reported that sNfL levels increased alongside the EDSS scores and age and remained high for a longer time after relapse in patients with NMOSD, in line with our finding (16). Furthermore, our team also investigated sNfL level of Chinese NMOSD patients with single molecule array (SIMOA) method, a more sensitive assay, and the level of sNfL was lower [medium 7.97 (range 10.55–27.94) pg/mL] than that in this study (38). The deviation might attribute to the difference between these two cohorts. Taken together, sNfL could be a biomarker of disease activity and disability but might not influence the course of an NMOSD attack. It is noteworthy sNfL showed predictive value for long-term clinical outcomes in multiple sclerosis and Guillain-Barré syndrome (39, 40).

In conclusion, the outcome of attacks of NMOSD was generally moderate, and ON might have poorer remission rates than MY during 1 month. Furthermore, this study identifies Th2-related cytokines characterizing the prognosis of acute episodes of NMOSD at 1 month, and sNfL is likely to be a biomarker of disease severity at attack. It might extend the clinical knowledge of the treatment response and prognosis and severity of patients with NMOSD attacks. A limitation that should be noted is that COVID-19 raised the rate of loss to follow-up, which caused small numbers of cases in some subgroups. Meanwhile, we measured serum biomarkers by MSD other than SIMOA which previous study always applied. Although the concentration of serum biomarkers included NfL and cytokines/chemokines detected in our study was tend to be consistent with that detected by SIMOA in other studies (16, 39, 41, 42), prospective studies are warranted to confirm the current findings.

The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

The studies involving human participants were reviewed and approved by The Institutional Ethical Review Board of the Third Affiliated Hospital of Sun Yat-Sen University, and the Institutional Committee approved the experiments performed on patients (ID [2018] 02-362-02). The patients/participants provided their written informed consent to participate in this study.

JW and CC contributed to the data acquisition, drafting of the manuscript, statistical analysis, and technical and material support. YC, YW, YML, HW, ZW and RL were involved in technical support and participant enrolment. XS contributed to measure AQP4-IgG levels and multiple cytokines and chemokines in serum. YXL and YS contributed to the analysis and interpretation of data and technical support. ML and SH was involved in participant enrolment and technical and material support. ZL and WQ contributed to design and conceptualized study, interpreted the data, revised the manuscript for intellectual content. ST was involved in the study concept and design, critical revision of the manuscript, procurement of funding, study supervision, and final approval of the version to be published. All authors contributed to the article and approved the submitted version.

This study was funded by China Postdoctoral Science Foundation (2018M643335), the Fundamental Research Funds for the Central Universities (2021qntd34) and Guangdong Science and Technology Department-Regional Joint Fund (2020A1515111133).

Authors ML and SH were employed by company Harbour BioMed Therapeutics Limited.

The remaining authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.