Serum antinuclear antibodies associate with worse prognosis in AQP4‐positive neuromyelitis optica spectrum disorder

Abstract Background Studies have demonstrated that antinuclear antibodies (ANAs) may be correlated with neuromyelitis optica spectrum disorder (NMOSD). However, the conflicting results of studies about the value of ANAs in AQP4 antibody‐positive NMOSD patients need to be further investigated. Material Case data were collected from 143 patients with AQP4 antibody‐positive NMOSD. Patients were divided into two groups based on the ANA test results. The analysis of clinical characteristics, laboratory tests, and MRI examination results were compared between two groups: the NMOSD patients with ANA (+) and with ANA (−). Results Disease duration of NMOSD is shorter in the ANA (+) patients with EDSS < 4 than in the ANA (−) patients (12.05 ± 16.73 versus 29.43 ± 41.03, p‐value = .013). The median time from disease onset to an EDSS score of 4.0 is significantly longer in the ANA (−) NMOSD patients than in the ANA (+) patients (48.2 months versus 24 months, p = .04). In addition, ANA (RR, 2.234; 95% CI, 1.078–4.629; p‐value = .031) can predict the severity of NMOSD. Conclusions Antinuclear antibodies seem to be associated with more severe disease activity in NMOSD patients.

In recent years, increasing evidence has demonstrated that AQP4-IgG causes damage to peripheral organs beyond the CNS such as skeletal muscle, vestibulocochlear nerves, gastrointestinal tract, blood system, kidney, lung, and placenta (He et al., 2017).
Clinically, a considerable number of NMOSD patients are positive for other autoimmune antibodies including antinuclear antibodies (ANAs), which are produced against DNA, RNA, proteins, or their molecular complexes in the cell nucleus. ANAs can be characterized in many autoimmune diseases such as systemic lupus erythematosus (SLE) and determine the activity and prognosis of SLE (Egner, 2000;Pisetsky, 2017). It has been reported that AQP4-Ab was detectable in 84.6% patients with SLE and NMO or LETM/rON and in 62.5% patients with other CTD and NMO or LETM, which showed the coexisting relationship between other autoimmune diseases and AQP4-Ab-positive NMOSD patients (Jarius et al., 2011). ANAs were reported to be more frequently exist in NMO-IgG-seropositive patients (Pittock et al., 2008), while another report also showed that patients with AQP4(+) were more likely to have coexisting autoantibodies (45/97, 46,4%) and autoimmune disorders (31/130, 23.8%) (Jarius et al., 2012). However, it is still unknown whether ANAs have any clinical significance or represent an epiphenomenon in NMOSD.
A study showed that ANAs may be a protective marker in NMOSD.
In contrast, another study found that ANAs are not sufficient to indicate the severity of NMOSD disease (Lee et al., 2019;Masuda et al., 2016). The contradictory results of these studies motivate us to further investigate the role of ANA in NMOSD.
In this paper, we hypothesized that more severe disease activity may have the potential to be occurred in ANA (+) NMOSD patients, and we conducted to evaluate the value of ANAs in measuring disease severity and prognosis of patients with APQ4 antibody-positive NMOSD by comparing the clinical characteristics, laboratory tests, and MRI examination results of AQP4 (+) NMOSD patients with ANA (+) or ANA (−). The diagnosis of NMOSD was based on the 2015 international consensus diagnostic criteria (Wingerchuk et al., 2015). The disability status of NMOSD patients after admission was determined by the use of the Expanded Disability Status Scale (EDSS) before any treatments (Kurtzke, 1983) and classified as mild disability (EDSS score 0-3.5) or moderate/severe disability (EDSS score 4-9.5) (Conradsson et al., 2018;Sicras-Mainar et al., 2017). EDSS with score 4.0 (hereafter referred as to EDSS 4.0) is considered to be a key disability milestone (Harding et al., 2019). Thus, EDSS 4.0 was used as the cutoff point in this study. Other systemic autoimmune diseases were diagnosed by rheumatologists according to the established criteria (Aringer et al., 2019;Garber et al., 2012;Ross et al., 2016;Shiboski et al., 2017).

| Samples and study design
The MRI examination time of patients was within 5-7 days after admission at the disease onset period. Clinical characteristics and MRI findings during attacks also were compared between the groups.
About 63.5% (33/52) of patients with ANA (+) and 68.1% (62/91) of patients with ANA (−) were first diagnosed by our department, and no immunosuppressive drugs were used before the diagnosis and the ANA tests. The others had received immunosuppressive treatment in other hospitals during the stable period previously. Patients were recruited and tested at the acute phase, and ANA test was performed in the first day morning after admission. After blood samples of ANA test were collected, all NMOSD patients received at least one therapy such as methylprednisolone pulse therapy, intravenous immunoglobulin.
Patients were tested for AQP4 antibody using a cell-based assay from a commercial BIOCHIP kit (Euroimmun) (Long et al., 2012).
Serum samples were diluted to 1:9, and each sample was assayed at least twice, by two examiners blind to the origin of the specimens.
Samples with positive results twice were deemed to be positive. We also examined CSF parameters, including white blood cells (WBC) and total protein (TP).

| Statistical analyses
All the data in this study were presented as mean ± standard deviation (SD) or median ± interquartile range (IQR). Characteristics were compared using Pearson's chi-square test (or Fisher's exact) for categorical variables and two-sample t test (or nonparametric Mann-Whitney U test) for continuous variables. The survival time to an EDSS 4.0 was displayed using the Kaplan-Meier curves; log-rank test was used to compare the survival experience between the ANA (+) and ANA (−) groups. The prognosis value was analyzed by the Cox multivariate model. Due to the exploratory nature of the study, no adjustment for multiple comparisons was made.
All statistical analyses were performed by the Statistical Program for the Social Sciences (SPSS) statistical software (version 22.0). A two-sided p-value < .05 was considered statistically significant. There are no significant differences in age and sex distribution between ANA (+) patients and ANA (−) patients (p > .05). The disease duration in the ANA (+) patients is shorter compared with the ANA (−) patients (18.94 ± 24.48 months versus 30.14 ± 40.20 months, p = .044). Autoimmune diseases including Sjogren's syndrome (SS), rheumatoid arthritis (RA), SLE, and hyperthyroidism are confirmed in all of the NMOSD patients, and 28.85% ANA (+) patients are diagnosed with SS, which have a significantly higher incidence than ANA (−) patients (28.85% versus 4.40%, p = .001). The main symptoms at onset attack, EDSS, ARR, CSF WBC counts, and CSF TP level are not statistically significant with p-value larger than 0.05.

| D ISCUSS I ON
With the deepening research on NMOSD and its related AQP4 autoantibody, more attention has been attracted on the relation be-   (Ruffati et al., 1990). However, the mean age did not differ between ANA-positive and ANA-negative NMOSD patients, or between NMOSD patients with and without high levels of ANA. In some patients with symptoms of NMOSD, no AQP4-Abs but Abs against myelin-oligodendrocyte-glycoprotein (MOG) are detectable.

ANA (+) NMOSD ANA (−) NMOSD p-value ANA (+) NMOSD ANA (−) NMOSD p-value
These clinical syndromes are now frequently referred to as "MOGencephalomyelitis" (MOG-EM). Although the frequency of coexistent autoimmune diseases seems to be lower than AQP4-Ab-positive patients, comorbidity with other autoimmune disorders has been reported in MOG-EM patients (Borisow et al., 2018). Since all patients in this study were recruited as AQP4 antibody-positive NMOSD Chinese Han patients, the relationship between ANAs and MOG-EM is worth to follow in the future. Steroid-treated patients appeared to have a much lower frequency of ANA than non-steroid-treated patients (Pozzilli et al., 1998). The ANA test in this study was per- In this study, the clinical characteristics including the main symptoms of the onset attack, EDSS score, ARR, and CSF examinations do not differ between the ANA groups, which are consistent with previous study (Lee et al., 2019). Besides, we found the time from   and the time to reach EDSS score of 6.0 was shorter than that with negative ANAs (Masuda et al., 2016). With the EDSS score of 6.0 as the cutoff, we conducted the sensitivity analysis. We analyzed the time to the EDSS score of 6.0 using the Kaplan-Meier curves, but no significant difference was found. The number of patients in the above studies was smaller than ours and relied on EDSS score of 6.0 as the determination of disease severity, leading to the differences in results from what we got. The presence of ANA reflects ongoing autoimmune stimulation by the exposition of self-antigens during oligodendrocyte apoptosis, which is associated with certain pathological patterns of MS. Early apoptotic oligodendrocyte lesion has been postulated as an early event (Barnett & Sutton, 2006), and ANAs were detected mainly in patients with early diseases (Szmyrka-Kaczmarek et al., 2012), which means EDSS score of 4.0 is more commonly used as the reference of early and middle stage of inflammatory demyelinating disease of the human central nervous system. Therefore, in order to study the clinical predictive value of ANAs in NMOSD more accurately, EDSS 4.0 is reasonable to be selected as the cutoff point to evaluate the severity and prognosis of the disease.
According to the results we got, ANAs were observed to be associated with the poorer NMOSD prognosis. Another study showed the similar results that primary and secondary progressive disease had a higher antibody frequency than relapsing-remitting (p < .05) or benign (p < .001) MS (Spadaro et al., 1999). Potential mechanisms have been hypothesized that a generalized immune dysregulation occurred, involving activation of both autoreactive Th1 cells (mainly linked to CNS lesions) and B cells via Th2 cells (Cerutti et al., 2005;Desai et al., 1993). Besides, ANAs could cause inflammation and tissue damage through binding to exposed chromatin fragments or directly to intrinsic antigens by cross-reactivity, or deposing in tissue lesions by forming immune complexes of antibodies with DNA or nucleosomes (Fenton et al., 2009;Rekvig, 2015;Seredkina et al., 2013).

| Limitations
With the advantages of large simple size and prognostic analysis of is true that there are some patients who had received immunosuppressive treatment or IVIG treatment in other hospitals during the stable period previously, and the results of ANA test were very hard to be collected at the time of their initial onset. Therefore, we did not conduct long-term follow-up on ANA status, which is also a limitation of this study. Further study is needed about the relationship among different treatments, ANA status changes, and EDSS scores.

| CON CLUS IONS
In conclusion, our results suggested that ANA seems to be more associated with the severe disease activity changes in AQP4 antibodypositive NMOSD patients, which further implied that ANA can be able to use as a prognostic marker in NMOSD.

CO N FLI C T O F I NTE R E S T
All authors declare that there are no conflicts of interest.

AUTH O R S ' CO NTR I B UTI O N S
Y. J. and F. P. contributed to the conception and design of this study.
F., Y. Z., Y. X., J. T., Y. C., F. P., and Y.J. drafted the manuscript. All the authors read and approved the final manuscript.

E TH I C A L A PPROVA L
The study was conducted according to the principles expressed in the Declaration of Helsinki and approved by the Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University (No.2017-33). All study participants gave written informed consent for research and publication.

PEER R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1002/brb3.1865.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.