Should Aquaporin-4 Antibody Test Be Performed in all Patients With Isolated Optic Neuritis?

Optic neuritis (ON) may be the initial manifestation of neuromyelitis optica spectrum disorder (NMOSD). Aquaporin-4 antibody (AQP4 Ab) is used to diagnose NMOSD. This has implications on prognosis and is important for optimal management. We aim to evaluate if clinical features can distinguish AQP4 Ab seropositive and seronegative ON patients.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Should Aquaporin-4 Antibody Test Be Performed in all Patients With Isolated Optic Neuritis? Rosalynn G. Siantar, MBBS, MMed, FRCOphth, FAMS, Farah N.I. Ibrahim, MBBS, MMed, Hla M. Htoon, BDS, PhD, Sharon L.C. Tow, MBBS, FRCS(Ed), FAMS, Kong Yong Goh, MBBS, MMed, FRCS(Ed), FRCOphth, FAMS, Jing Liang Loo, MBBS, MMed, FRCS(Ed), FAMS, Su Ann Lim, MBBS, MMed, FRCS(Ed), Dan Milea, MD, PhD, Melissa C.H. Tien, MBBS, MMed, FAMS, Zhiyong Chen, MBBS, MRCP, MMed, Tianrong Yeo, MBBS, MRCP, MMed, FAMS, MCI, Josiah Y.H. Chai, MBBS, MRCP, FRCP, FAMS, Shweta Singhal, MBBS, MMed, FRCOphth, FAMS, PhD, Chee Fang Chin, MBChB, MMed, MRCS(Ed), FRCS(Ed), FAMS, Kevin Tan, BMedSci, BMBS, MRCP, FAMS, MS-HPEd Background: Optic neuritis (ON) may be the initial manifestation of neuromyelitis optica spectrum disorder (NMOSD). Aquaporin-4 antibody (AQP4 Ab) is used to diagnose NMOSD. This has implications on prognosis and is important for optimal management. We aim to evaluate if clinical features can distinguish AQP4 Ab seropositive and seronegative ON patients. Methods: We reviewed patients with first episode of isolated ON from Tan Tock Seng Hospital and Singapore National Eye Centre who tested for AQP4 Ab from 2008 to 2017. Demographic and clinical data were compared between seropositive and seronegative patients. Results: Among 106 patients (120 eyes) with first episode of isolated ON, 23 (26 eyes; 22%) were AQP4 Ab positive and 83 (94 eyes; 78%) were AQP4 Ab negative. At presentation, AQP4 Ab positive patients had older mean onset age (47.9 ± 13.6 vs 36.8 ± 12.6 years, P , 0.001), worse nadir VA (OR 1.714; 95% CI, 1.36 to 2.16; P , 0.001), less National Healthcare Group Eye Institute (RGS, SAL, MT, CFC), Tan Tock Seng Hospital, Singapore; Department of NeuroOphthalmology (FI, ST, JLL, DM, SS), Singapore National Eye Centre, Singapore; Data Science Unit (HMH), Singapore Eye Research Institute, Singapore; Mount Elizabeth Novena Hospital (KYG), Singapore; Department of Neurology (ZC, TY, JC, KT), National Neuroscience Institute, Singapore; and Duke-NUS Medical School (KT), Singapore K. Tan has received travel grants and compensation from Novartis, Merck, Sanofi, Eisai, and Viela Bio for consulting services. The remaining authors report no conflicts of interest. R.G. Siantar and F. Ibrahim contributed equally as first authors of the manuscript. C.F. Chin and K. Tan contributed equally as senior authors of manuscript. Address correspondence to Shweta Singhal, MBBS, MMed, FRCOphth, FAMS, PhD, Department of Neuro-Ophthalmology, Singapore National Eye Centre, 11 Third Hospital Avenue, Singapore 168751; E-mail: [e-mail] 454 optic disc swelling (OR 5.04; 95% CI, 1.682 to 15.073; p = 0.004), and higher proportions of concomitant anti-Ro antibody (17% vs 4%, p = 0.038) and anti-La antibody (17% vs 1%, p = 0.008). More AQP4 Ab positive patients received steroid-sparing immunosuppressants (74% vs 19%, p , 0.001) and plasma exchange (13% vs 0%, p = 0.009). AQP4 Ab positive patients had worse mean logMAR VA (visual acuity) at 12 months (0.70 ± 0.3 vs 0.29 ± 0.5, p = 0.051) and 36 months (0.37±0.4 vs 0.14 ± 0.2, p = 0.048) follow-up. Conclusion: Other than older onset age and retrobulbar optic neuritis, clinical features are non-discriminatory for NMOSD. We propose a low threshold for AQP4 Ab serology testing in inflammatory ON patients, particularly in high NMOSD prevalence populations, to minimize diagnostic and treatment delays. Journal of Neuro-Ophthalmology 2022;42:454–461 doi: 10.1097/WNO.0000000000001573 © 2022 by North American Neuro-Ophthalmology Society O ptic neuritis (ON) is an immune-mediated inflammatory condition of the optic nerve that leads to subacute or acute visual loss. ON may be idiopathic or because of secondary causes such as autoimmune diseases (1). ON may be the initial manifestation of neuromyelitis optica spectrum disorder (NMOSD), an inflammatory disease of the central nervous system (CNS), which includes aquaporin-4 antibody (AQP4 Ab) autoimmune astrocytopathic disease (2,3). AQP4 Ab is used to diagnose NMOSD; early diagnosis has important implications in prognosis and clinical management (4, 5). There is increasing evidence that NMOSD patients benefit from early treatment with highdose intravenous methylprednisolone or plasma exchange Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution during attacks and long-term immunosuppression to prevent attacks (6–8). However, AQP4 Ab serology is often unavailable in resource poor settings or only available in more specialized hospitals (9). Physicians managing de novo patients need to make initial clinical management decisions without knowledge of serostatus. Previous studies have attempted to identify key clinical and demographic characteristics distinguishing AQP4 Ab seropositive ON from seronegative ON. AQP4 Ab seropositivity is associated with severe and treatment refractory ON in Caucasian and Asian cohorts; relapse rates and visual outcomes were variable (10–19). Seropositive patients are more often women (10–14,18,20) and more likely to have co-existing autoantibodies (antinuclear antibody [ANA] and anti-Ro and anti-La antibodies) (10–13). No difference in ON onset age was found in Chinese (12) and Caucasian populations (4,10), whereas a trend toward increasing age with seropositivity was seen among Japanese patients (11,14,17). Our study aimed to evaluate the differences between AQP4 Ab seropositive and seronegative ON in a multiethnic Singapore-based cohort of patients and identify clinical features of ON that may predict AQP4 Ab seropositivity. METHODS Study Population We reviewed all ON patients from Tan Tock Seng Hospital and Singapore National Eye Centre, tested for AQP4 Ab in the National Neuroscience Institute from August 2008 to December 2017. Patients presenting with first episode of isolated ON were included. Patients with known history of NMOSD or other manifestations of NMOSD were excluded. Serum samples were collected from each subject and assayed for AQP4 Ab by methods described below. This study follows the principles of the Declaration of Helsinki with ethical approval obtained from SingHealth Centralized Institutional Review Board (Reference number: 2016/2,201). Antibody Assays From August 2008 to October 2011, AQP4 Ab testing was performed using a live cell-based assay by immunofluorescence using green fluorescent protein-AQP4 fusion proteintransfected human embryonic kidney-293T cells (research protocol, Kyushu University, Japan). From November 2011 onward, AQP4 Ab testing was performed using a commercial fixed cell-based assay (Euroimmun, Lübeck, Germany) in the National Neuroscience Institute, Singapore. records. Detailed neuro-ophthalmic examinations, including visual acuity (VA), Ishihara color vision, Humphrey visual field (mean deviation and pattern SD), and manifest refraction where performed, were recorded. The objective lowest VA recorded in the observational period was defined as nadir VA. Clinical examination included slit-lamp examination and fundus biomicroscopy. Serum samples were tested for AQP4 Ab, ANA, anti–double-stranded DNA, anti-Ro and anti-La antibodies for select cases. Patients underwent MRI of the brain and orbits. Patients were followed up for up to 36 months. Statistical Analysis VA was presented in logarithm of the minimum angle of resolution (logMAR) units; counting fingers was converted to 2.6 logMAR, hand movements was converted to 2.9 logMAR, light perception was converted to 3.1 logMAR, and no light perception was converted to 3.4 logMAR (21). All statistical analyses were performed using IBM Corp. Released 2016. IBM SPSS Statistics for Windows, Version 24.0. Armonk, NY: IBM Corp. Comparative analyses between groups were performed using Mann–Whitney U test for continuous variables, and Chi-square/Fisher’s Exact tests for categorical variables. P value for significance was set at ,0.05. Generalized estimating equation (GEE) was used in the regression models to compare AQP4 Ab status as the dependent variable and adjusted for both eyes at each time point. Univariate GEE analysis was also performed to compare AQP4 Ab status as the dependent variable and adjusted for both eyes, all visit timepoints, age at presentation, VA, disc swelling, and time to nadir. RESULTS A total of 258 patients with ON sought treatment in the 2 institutions; 106 patients had AQP4 Ab testing performed and were included in the analysis (Fig. 1). Among 106 patients (120 eyes) with first episode of isolated ON, 23 (26 eyes; 22%) were AQP4-Ab-positive and 83 (94 eyes; 78%) were AQP4-Ab-negative. Table 1 summarizes the characteristics of study participants. Patient Demographics AQP4-Ab-positive patients had older mean onset age (47.9 ± 13.6 vs 36.8 ± 12.6 years, P , 0.001), highest proportion of Chinese and lower proportions of Malay, Indian, and other races (20:3:0:0 vs 44:8:13:18, P = 0.006). There was a higher proportion of women in the AQP4 Ab seropositive group compared with the seronegative group, which was not statistically significant (82.6% vs 69.9%, P = 0.31). Demographic, Clinical, Laboratory, and Radiological Parameters Clinical Characteristics According to Aquaporin4 Antibody Serostatus Age, gender, ethnicity, presenting symptoms, and treatment regimen were recorded through manual review of case Clinical characteristics of study participants with ON are described in Tables 2 and 3. AQP4-Ab-positive patients Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 455 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution (Table 5). Worse nadir VA (odds ratio [OR] 1.714; 95% confidence interval [CI], 1.36 to 2.16; P , 0.001) and no optic disc swelling (OR 5.04; 95% CI, 1.682 to 15.073; P = 0.004) were more likely to be AQP4-Ab-positive. CONCLUSIONS FIG. 1. Flow chart of patient recruitment. were less likely to have optic disc swelling (19% vs 55%, P = 0.001) compared with AQP4-Ab-negative patients. There were no differences in presenting VA, time from onset to nadir, and other presenting visual functions. AQP4-Ab-positive patients had higher frequencies of concomitant anti-Ro antibody (17% vs 4%, P = 0.038) and anti-La antibody (17% vs 1%, P = 0.008). More AQP4-Ab-positive patients received steroid-sparing immunosuppressants (74% vs 19%, P , 0.001) and plasma exchange (13% vs 0%, P = 0.009) treatments. There were no differences in duration of symptoms, presence of pain, unilateral vs bilateral presentation or MRI detection of optic nerve enhancement, and chiasmal involvement. Post-treatment Visual Acuity Table 4 describes the post-treatment visual functions adjusted for both eyes assessed at each cross-sectional time point. Mean VA was worse in the AQP4-Ab-positive patients at 12 months (0.70 ± 0.3 vs 0.29 ± 0.5, P = 0.051) and 36 months (0.37 ± 0.4 vs 0.14 ± 0.2, P = 0.048). However, there were no differences in posttreatment mean best VA and mean VA at last recorded follow-up in AQP4-Ab-positive and negative groups. Generalized Estimating Equation Analysis Univariate GEE analysis comparing AQP4 Ab serostatus as the dependent variable was performed adjusting for both eyes and repeated measures of VA at each time point In this study, we evaluated the clinical features of patients with ON in a multiethnic cohort in Singapore dichotomized by AQP4 Ab serostatus. The frequency of seropositive patients in our study was 22%. Other Asian cohorts have reported variable frequencies—32.6% Chinese (18), 12% Japanese (11), and 14% Korean (22). Although Jarius et al reported a higher frequency in the Caucasian population (78.3%) (10), this could be explained by a highly selected study cohort with the inclusion of bilateral ON, recurrent ON, and patients with prior history of NMOSD. In our cohort, the proportion of female patients was higher in the seropositive group, but this was not statistically significant. Previous studies have reported a strong female preponderance in AQP4 Ab ON patients (4,12,20,23). Female gender has been reported as a risk factor for CNS demyelinating diseases, including NMOSD (24). The exact mechanisms of gender differences in autoimmunity are incompletely understood, but may be mediated by hormonal and genetic effects (24). In our study, there was highest proportion of Chinese AQP4-Ab-positive patients and lower proportions of Malay, Indian, and other races. Racial differences may play a role in the prevalence of NMOSD (2). However, previous studies characterizing AQP4 Ab seropositive patients in Europe or Asia have been limited to single ethnic groups (4,5,10– 17,19). In a recent review of CNS inflammatory demyelinating diseases in Singapore, Tan and colleagues reported that compared with other races, the proportion of seropositive patients was highest in Chinese. AQP4-Ab-positive NMOSD prevalence is highest in Chinese (4.5 per 100,000), followed by Eurasian (mixed Caucasian-Asian ancestry) (3.9 per 100,000), Malay (1.5 per 100,000), and Indian (1.1 per 100,000) (25). Therefore, we propose that AQP4 Ab serology should be tested in every patient TABLE 1. Baseline demographics of patients Characteristics AQP4-Ab-Positive (n = 23) Age at onset, yr (±SD) Gender (female), n (%) Race, n (%) Chinese Malay Indian Others 47.87 (13.6) 19 (82.6) 20 (87) 3 (13) 0 0 AQP4-Ab-Negative (n = 83) P* 36.77 (12.6) 58 (69.9) ,0.001 0.31 44 8 13 18 (53) (9.6) (15.7) (21.7) 0.006 Data presented are mean (SD) or frequency (percentage), where appropriate. *P-value was based on Mann–Whitney U Test and Fisher’s Exact or chi-square test where appropriate. AQP4 Ab, Aquaporin-4 antibody. 456 Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Table 2. Clinical characteristics of patients with optic neuritis (by person) Characteristics Age at onset; yr mean(±SD) Duration of symptoms; days mean (±SD) Ocular pain, yes n (%) Affected eye n (%) Unilateral Bilateral Imaging Optic nerve enhancement, yes n (%) Unilateral Bilateral Chiasmal involvement, yes Co-existing autoantibodies n (%) Anti-MOG+ ANA+ Anti-dsDNA+ Anti-Ro+ Anti-La+ Treatment n (%) Pulsed + oral steroids Steroid-sparing immunosuppressants Plasmapheresis AQP4-Ab-Positive (n = 23) AQP4-Ab-Negative (n = 83) P* 47.87 (13.6) n = 21 7.5 (5.0) 10 (43.5) 36.77 (12.6) n = 79 7.1 (7.9) 46 (55.4) ,0.001 0.401 20 (87) 3 (13) 72 (86.7) 11 (13.3) 1 15 (65.2) 3 (13) 4 (17.4) 53 (63.9) 6 (7.4) 6 (7.2) 1 0.404 0.219 n=1 0 n = 21 5 (21.7) n = 22 2 (8.7) n = 21 4 (17.4) n = 21 4 (17.4) n=9 2 (2.4) n = 79 15 (19.3) n = 77 2 (2.4) n = 73 3 (3.6) n = 73 1 (1.2) 1 22 (95.7) 17 (73.9) 3 (13) 73 (88) 16 (19.3) 0 0.31 1 0.205 0.038 0.008 0.45 ,0.001 0.009 Data presented are mean (SD) or frequency (percentage), where appropriate. *P-value was based on Mann Whitney U test and Fisher’s Exact or chi-square test where appropriate. AQP4 Ab, Aquaporin-4 antibody; Anti-MOG, antimyelin oligodendrocyte glycoprotein antibodies; ANA, antinuclear antibody; Anti-dsDNA, anti–double-stranded DNA antibody. presenting with inflammatory ON in the Chinese ethnic group where this is a relatively high prevalence of NMOSD, and recommend a low threshold for AQP4 Ab testing in patients of other ethnicities. This would ensure timeliness in the diagnosis of NMOSD. Our AQP4 Ab seropositive patient cohort were a decade older at disease onset (47.9 years). The typical age of presentation for NMOSD ON is 35–45 years (26). Jarius et al reported that the median age of onset in the AQP4 Ab seropositive patients was higher than that in classical multiple sclerosis (10). Studies from Japan and China reported similarly older age of onset in AQP4-Ab-positive ON (14,17,27). No difference in presenting VA or laterality of involvement was found in our cohort between the seropositive and seronegative groups. Although a previous study in Caucasian patients found that seropositive patients had poorer VA during acute attacks (10), it is difficult to predict the presence or absence of AQP4 Ab based on VA at presentation alone (11). Previous reports of laterality of involvement have been directly contradic- Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 tory with bilateral involvement being associated with seropositive (28) and seronegative patients (4,10) in different studies. We found that patients with worse nadir VA were more likely to be AQP4-Ab-positive. This is in line with the notion that NMOSD is typically characterized by more severe ON attacks, and that VA loss at nadir is more severe in NMOSD compared to multiple sclerosis (MS) (3). This is also supported by Ciotti et al and Masuda et al who reported worse nadir VA in NMOSD compared with MS during the natural course of an ON attack (29,30). The latter also found that time to nadir is earlier in NMOSD than in MS, although we did not find a difference between the AQP4-Ab-positive and negative groups. In agreement with Seay and Rucker (31), we propose that severe visual loss at nadir should prompt definite testing for AQP4 Ab. A key finding in our study was the negative correlation of disc swelling (significant OR of 5.04) with AQP4 Ab seropositivity. This is consistent with current understanding that inflammation in AQP4-Ab-positive ON usually affects the posterior/pre-chiasmatic portion of the optic nerve 457 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Table 3. Clinical characteristics of patients with optic neuritis (by eye) Characteristics AQP4-Ab-Positive AQP4-Ab-Negative Presenting VA (logMAR) of affected eye Median (interquartile range) #0.5 0.5–CF CF or worse Nadir VA (logMAR) of affected eye Median (interquartile range) #0.5 0.5–CF CF or worse Time from onset to nadir (d) in affected eye mean (±SD) n = 26 0.96 (0.18–2.9) 12 (46.2) 2 (7.6) 12 (46.2) n = 26 2.6 (0.5–3.2) 6 (23.1) 5 (19.2) 15 (57.7) n = 21 9.9 (5.6) n = 25 5.24 (6.9) n = 16 216.1 (11.8) 6.3 (4.9) n = 20 21.49 (3.0) n = 26 5 (19.2) n = 94 0.75 (0.18–2.6) 41 (43.6) 25 (26.6) 28 (29.8) n = 90 1.3 (0.5–2.6) 21 (23.3) 30 (33.3) 39 (43.3) n = 77 8.9 (6.7) n = 92 6.8 (6.7) n = 75 214.7 (9.9) 7.5 (3.9) n = 52 21.5 (2.7) n = 94 52 (55.3) Ishihara color vision (/15) mean (±SD) of affected eye Humphrey visual field mean (±SD) of affected eye Mean deviation Pattern SD Spherical error (diopters) mean (±SD) of affected eye Disc swelling, yes n (%) in of affected eye P* 0.236 0.086 0.18 0.329 0.348 0.369 0.88 0.298 1.0 0.001 Data presented are mean (SD) or frequency (percentage), where appropriate. *P-value was based on Mann–Whitney U test and Fisher’s Exact or chi-square test where appropriate. AQP4 Ab, Aquaporin-4 antibody; CF, counting fingers; LogMAR, logarithm of the minimum angle of resolution; VA, visual acuity. causing retrobulbar ON (11). Previous MRI-based studies have also established that NMOSD ON lesions are typically longitudinal, extending from the posterior globe to the optic tract, whereas MS lesions are more focal and localized in anterior intra-orbital segment (32). In agreement with previous reports, we found high frequency of anti-Ro and anti-La antibody positivity in our AQP4 Ab seropositive group (11). NMO has been reported to be the first manifestation of systemic lupus erythematosus (SLE), with AQP4 Ab existing in the sera for years before the first NMO attack in patients with SLE (33). Visual acuity outcomes in our seropositive cohort were poorer with worse visual outcomes in the follow-up period at 12 and 36 months compared with the seronegative group. Previous reports of visual recovery after ON revealed mixed results, with no significant difference between seropositive and seronegative groups in some cases (4), and poorer outcomes in AQP4 Ab ON compared with MS in others (16,30). Similar proportions of patients with typical and atypical ON test positive for AQP4 Ab (34). Multiple groups across the world have attempted to identify the clinical and demographic features of AQP4 Ab ON that aid in guiding serological testing for AQP4 Ab. Our data taken together with these affirm that none of these features show consistent correlation in predicting AQP4 Ab positivity. Although routine testing for AQP4 Ab was considered controversial in the past (31), given the worse prognosis and need for emergent treatment of relapses, and immunosup458 pressive treatment for better long-term outcomes, the early identification of NMOSD is critical (3). This would be more important in the Asian, African, and eastern Mediterranean populations where the prevalence of AQP4 Ab ON is greater than in the Caucasian population (2). However, less than 15% of the lower income countries (home to over 90% of the high-risk populations listed above) currently have access to AQP4 Ab serological testing (9). To enable optimal management for best visual outcomes, we recommend greater efforts to improve access to AQP4 Ab testing globally. Our study has several limitations. First, only patients who were tested for AQP4 Ab were included. This was performed with the intention of having a cohort with clearly defined serostatus to allow for robust comparison. However, because the study design was retrospective in nature, testing was based on clinician discretion and not systemically performed for all patients. It is possible that clinical features in cases selected for AQP4 Ab testing may have resulted in selection bias. Second, despite recruiting patients from 2 tertiary Ophthalmology referral centers, our sample size is relatively small and from a single country. Third, the AQP4 Ab was not all measured at the time of acute ON; some were tested after recovery. As AQP4 Ab levels fluctuate and may decrease with immunotherapy, this may underestimate the actual proportion of seropositive cases. Fourth, there was heterogeneity in the type, timing, and duration of treatment regimens among individual clinicians, which may have influenced the final outcomes. Fifth, we did not have specific data Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Table 4. Post-treatment visual functions of patients with optic neuritis, adjusted for both eyes using Generalized Estimating Equation (by eye) AQP4-Ab-Positive AQP4-Ab-Negative P* n = 25 0.90 (1.3) n = 23 8.8 (7.7) n = 18 27.2 (6.8) 5.1 (3.8) n = 73 0.42 (0.8) n = 69 11.6 (6.7) n = 56 27.5 (8.9) 4.5 (3.0) 0.077 n = 18 0.62 (0.9) n = 17 11.2 (5.8) n = 13 28.8 (7.6) 5.4 (3.0) n = 51 0.49 (0.7) n = 49 11.8 (6.3) n = 34 27.8 (7.6) 4.8 (4.7) n = 19 0.70 (0.3) n = 19 8.9 (7.4) n = 16 27.3 (6.4) n = 16 5.0 (3.6) n = 55 0.29 (0.5) n = 54 11.8 (6.3) n = 37 27.1 (9.1) n = 36 5.1 (3.6) n = 17 0.48 (0.7) n = 16 8.7 (7.2) n = 12 26.3 (5.9) 5.1 (4.2) n = 37 0.17 (0.2) n = 36 13.1 (4.8) n = 23 26.6 (9.1) 4.2 (3.4) n = 14 0.37 (0.4) 9.2 (7.9) n = 12 23.7 (3.5) 3.4 (3.1) n = 25 0.65 (1.1) ref:1 0.432 OR (95% CI 0.03–5.4) 0.324 OR (95% CI 0.05–2.3) n = 25 0.76 (0.94) ref:1 0.127 OR (95% CI 0.025–0.661) 0.212 OR (95% CI 0.025–1.772) n = 31 0.14 (0.2) 13.6 (3.3) n = 20 24.3 (4.0) 3.8 (2.7) n = 80 0.29 (0.63) Visual Functions Post-Treatment At 3 mo follow-up VA (logMAR) mean (±SD) Ishihara color vision (/15) mean (±SD) Humphrey visual field, mean (±SD) Mean deviation Pattern SD At 6 mo follow-up VA (logMAR) Mean (±SD) Ishihara color vision (/15) mean (±SD) Humphrey visual field, mean (±SD) Mean deviation Pattern SD At 12 mo follow-up VA (logMAR) Mean (±SD) Ishihara color vision (/15) mean (±SD) Humphrey visual field, mean (±SD) Mean deviation Pattern SD At 24 mo follow-up VA (logMAR) Mean (±SD) Ishihara color vision (/15) mean (±SD) Humphrey visual field, mean (±SD) Mean deviation Pattern SD At 36 mo follow-up VA (logMAR) Mean (±SD) Ishihara color vision (/15) mean (±SD) Humphrey visual field, mean (±SD) Mean deviation Pattern SD Post-treatment best VA (logMAR) Mean (±SD) #0.5 0.5–CF CF or worse VA at last recorded follow-up (logMAR) Mean (±SD) #0.5 0.5–CF CF or worse 0.111 0.899 0.572 0.596 0.726 0.683 0.638 0.051 0.146 0.959 0.927 0.072 0.022 0.907 0.518 0.048 0.040 0.647 0.748 0.142 0.514 0.257 n = 80 0.35 (0.63) 0.098 0.014 0.152 Data presented are mean (standard deviation) or frequency (percentage), where appropriate. *P-value was based on Mann–Whitney U test and Fisher’s Exact or chi-square test where appropriate. AQP4 Ab, Aquaporin-4 antibody; CF, counting fingers; CI, confidence interval; logMAR, logarithm of the minimum angle of resolution; n, number of eyes; OR, odds ratio; VA, Visual acuity. pertaining to neuroimaging features of MS/NMOSD that could have provided additional useful information, especially for the patients who neither had optic nerve Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 enhancement nor chiasmal involvement. It would also be interesting to see whether patients were subsequently diagnosed with transverse myelitis, MS or NMOSD 459 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution Table 5. Univariate GEE analysis comparing AQP4 Ab serostatus as the dependent variable adjusted for both eyes and repeated measures at each time point 95% CI for OR Parameter VA at different timepoints 36 mo 24 mo 12 mo 6 mo 3 mo At nadir At presentation Absence of disc swelling VA, mean (of all timepoints) Age at presentation Time to Nadir OR Lower Upper P 0.355 0.370 0.452 0.482 0.501 1.714 Reference:1 5.04 0.662 0.940 0.984 0.302 0.270 0.313 0.346 0.370 1.36 0.482 0.507 0.652 0.672 0.680 2.16 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 ,0.001 1.682 0.454 0.894 0.960 15.073 0.965 0.988 1.008 0.004 0.032 0.015 0.193 AQP4 Ab, Aquaporin-4 antibody; GEE, Generalized Estimating Equation; CI, confidence interval; OR, odds ratio; VA, Visual acuity. during the follow-period. Although some patients were followed up for 36 months, there was a fairly high proportion of patients who were lost to follow-up. Finally, 2 different AQP4 Ab assays were used in our study—a live cell-based assay before November 2011, and a fixed cellbased assay thereafter. It is uncertain whether differences in the sensitivity and specificity of each assay could have had an impact on antibody detection. Limitations notwithstanding, our data suggest that differences in clinical features of ON may not be helpful in the selection of patients for AQP4 Ab testing. Our findings confirmed data from other studies in South East and East Asia that AQP4 Ab ON patients have poorer visual outcomes. Older patients and those with worse nadir VA and retrobulbar ON are more likely to have AQP4 Ab ON. However, these clinical differences alone are insufficient to reliably predict AQP4 Ab seropositivity. To ensure optimal management of patients with NMOSD who have poorer visual prognosis, we propose that routine testing for AQP4 Ab serology should be performed in all patients presenting with a first episode of isolated ON in the Chinese ethnic group where this is a relatively high prevalence of NMOSD, and recommend a low threshold for AQP4 Ab testing in patients of other ethnicities. STATEMENT OF AUTHORSHIP Conception and design: K. Tan, C. F. Chin, S. Singhal, K. Y. Goh; Acquisition of data: R. G. Siantar F. Ibrahim; Analysis and interpretation of data: H. M. Htoon, R. G. Siantar, F. Ibrahim, K. Tan, S. Singhal, C. F. Chin. Drafting the manuscript: R. G. Siantar, F. Ibrahim, S. Singhal, K. Tan, C. F. Chin; Revising it for intellectual content: K. Y. Goh, S. Tow, J. L. Loo, D. Milea, M. Tien, S. A. Lim, J. Chai, T. Yeo, Z. Chen. Final approval of the completed manuscript: K. Tan, C. F. Chin, S. Singhal. 460 ACKNOWLEDGMENTS From August 2008 to October 2011, AQP4 Ab testing was performed using a live cell-based assay by immunofluorescence using green fluorescent protein (GFP)-AQP4 fusion protein-transfected human embryonic kidney (HEK)293T cells (research protocol, Kyushu University, Japan). REFERENCES 1. Bennett JL. Optic neuritis. Continuum (Minneap Minn). 2019;25:1236–1264. 2. Hor JY, Asgari N, Nakashima I, Broadley SA, Leite MI, Kissani N, Jacob A, Marignier R, Weinshenker BG, Paul F, Pittock SJ, Palace J, Wingerchuk DM, Behne JM, Yeaman MR, Fujihara K. Epidemiology of neuromyelitis optica spectrum disorder and its prevalence and incidence worldwide. Front Neurol. 2020;11:501. 3. Chen JJ, Pittock SJ, Flanagan EP, Lennon VA, Bhatti MT. Optic neuritis in the era of biomarkers. Surv Ophthalmol. 2020;65:12–17. 4. Jarius S, Frederikson J, Waters P, Paul F, Akman-Demir G, Marignier R, Franciotta D, Ruprecht K, Kuenz B, Rommer P, Kristoferitsch W, Wildemann B, Vincent A. Frequency and prognostic impact of antibodies to aquaporin-4 in patients with optic neuritis. J Neurol Sci. 2010;298:158–162. 5. Petzold A, Pittock S, Lennon V, Maggiore C, Weinshenker BG, Plant GT. Neuromyelitis optica-IgG (aquaporin-4) autoantibodies in immune mediated optic neuritis. J Neurol Neurosurg Psychiatry. 2010;81:109–111. 6. Bonnan M, Valentino R, Debeugny S, Merle H, Fergé JL, Mehdaoui H, Cabre P. Short delay to initiate plasma exchange is the strongest predictor of outcome in severe attacks of NMO spectrum disorders. J Neurol Neurosurg Psychiatry. 2018;89:346–351. 7. Bonnan M, Cabre P. Plasma exchange in severe attacks of neuromyelitis optica. Mult Scler Int. 2012;2012:787630. 8. Levy M, Fujihara K, Palace J. New therapies for neuromyelitis optica spectrum disorder. Lancet Neurol. 2021;20:60–67. 9. Holroyd K, Vogel A, Lynch K, Gazdag B, Voghel M, Alakel N, Patenaude BN, Chiong-Rivero H, Mateen FJ. Neuromyelitis optica testing and treatment: availability and affordability in 60 countries. Mult Scler Relat Disord. 2019;33:44–50. Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 10. Jarius S, Ruprecht K, Wildemann B, Kuempfel T, Ringelstein M, Geis C, Kleiter I, Kleinschnitz C, Berthele A, Brettschneider J, Hellwig K, Hemmer B, Linker RA, Lauda F, Mayer CA, Tumani H, Melms A, Trebst C, Stangel M, Marziniak M, Hoffmann F, Schippling S, Faiss JH, Neuhaus O, Ettrich B, Zentner C, Guthke K, Hofstadt-van Oy U, Reuss R, Pellkofer H, Ziemann U, Kern P, Wandinger KP, Bergh FT, Boettcher T, Langel S, Liebetrau M, Rommer PS, Niehaus S, Münch C, Winkelmann A, Zettl U UK, Metz I, Veauthier C, Sieb JP, Wilke C, Hartung HP, Aktas O, Paul F. Contrasting disease patterns in seropositive and seronegative neuromyelitis optica: a multicentre study of 175 patients. J Neuroinflammation. 2012;9:14. 11. Ishikawa H, Kezuka T, Shikishima K, Yamagami A, Hiraoka M, Chuman H, Nakamura M, Hoshi K, Goseki T, Mashimo K, Mimura O, Yoshitomi T, Tanaka K. On behalf of the working group on diagnostic criteria for refractory optic neuritis based on neuroimmunological perspective epidemiologic and clinical characteristics of optic neuritis in Japan. Ophthalmology. 2019;126:1385–1398. 12. Li H, Wang Y, Xu Q, Zhang A, Zhou H, Zhao S, Kang H, Peng C, Cao S, Wei S. Features of anti-aquaporin 4 antibodyseropositive Chinese patients with neuromyelitis optica spectrum optic neuritis. J Neurol. 2015;262:2293–2304. 13. Ambika S, Balasubramanian M, Theresa L, Veeraputhiran A, Arjundas D. Aquaporin 4 antibody [NMO Ab] status in patients with severe optic neuritis in India. Int Ophthalmol. 2015;35:801–806. 14. Akaishi T, Nakashima I, Takeshita T, Kaneko K, Mugikura S, Sato DK, Takahashi T, Nakazawa T, Aoki M, Fujihara K. Different etiologies and prognoses of optic neuritis in demyelinating diseases. J Neuroimmunol. 2016;299:152– 157. 15. Jitprapaikulsan J, Chen JJ, Flanagan EP, Tobin WO, Fryer JP, Weinshenker BG, McKeon A, Lennon VA, Leavitt JA, Tillema JM, Lucchinetti C, Keegan BM, Kantarci O, Khanna C, Jenkins SM, Spears GM, Sagan J, Pittock SJ. Aquaporin-4 and myelin oligodendrocyte glycoprotein autoantibody status predict outcome of recurrent optic neuritis. Ophthalmology. 2018;125:1628–1637. 16. Sotirchos ES, Filippatou A, Fitzgerald KC, Salama S, Pardo S, Wang J, Ogbuokiri E, Cowley NJ, Pellegrini N, Murphy OC, Mealy MA, Prince JL, Levy M, Calabresi PA, Saidha S. Aquaporin-4 IgG seropositivity is associated with worse visual outcomes after optic neuritis than MOG-IgG seropositivity and multiple sclerosis, independent of macular ganglion cell layer thinning. Mult Scler. 2020;26:1360–1371. 17. Yamagami A, Wakakura M, Inoue K, Ishikawa H, Takahashi T, Tanaka K. Clinical characteristics of anti-aquaporin 4 antibody positive optic neuritis in Japan. Neuroophthalmology. 2019;43:71–80. 18. Kang H, Chen T, Li H, Xu Q, Cao S, Wei S. Prognostic factors and disease course in aquaporin-4 antibody-positive Chinese patients with acute optic neuritis. J Neurol. 2017;264:2130–2140. 19. Petzold A, Woodhall M, Khaleeli Z, Tobin WO, Pittock SJ, Weinshenker BG, Vincent A, Waters P, Plant GT. Aquaporin-4 and myelin oligodendrocyte glycoprotein antibodies in immunemediated optic neuritis at long-term follow-up. J Neurol Neurosurg Psychiatry. 2019;90:1021–1026. 20. Borisow N, Kleiter I, Gahlen A, Fischer K, Wernecke KD, Pache F, Ruprecht K, Havla J, Krumbholz M, Kümpfel T, Aktas O, Ringelstein M, Geis C, Kleinschnitz C, Berthele A, Hemmer B, Siantar et al: J Neuro-Ophthalmol 2022; 42: 454-461 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. Angstwurm K, Weissert R, Stellmann JP, Schuster S, Stangel M, Lauda F, Tumani H, Mayer C, Zeltner L, Ziemann U, Linker RA, Schwab M, Marziniak M, Then Bergh F, Hofstadt-van Oy U, Neuhaus O, Winkelmann A, Marouf W, Rückriem L, Faiss J, Wildemann B, Paul F, Jarius S, Trebst C, Hellwig K, Nemos. Influence of female sex and fertile age on neuromyelitis optica spectrum disorders. Mult Scler. 2017;23:1092–1103. Johnson LN, Guy ME, Krohel GB, Madsen RW. Levodopa may improve vision loss in recent-onset, nonarteritic anterior ischemic optic neuropathy. Ophthalmology. 2000;107:521– 526. Kim KM, Kim US. Aquaporin 4 antibody in the Korean patient with new-onset optic neuritis. Semin Ophthalmol. 2018;33:215–218. Kim SM, Waters P, Woodhall M, Kim YJ, Kim JA, Cheon SY, Lee S, Jo SR, Kim DG, Jung KC, Lee KW, Sung JJ, Park KS. Gender effect on neuromyelitis optica spectrum disorder with aquaporin4-immunoglobulin G. Mult Scler. 2017;23:1104– 1111. Billi AC, Kahlenberg JM, Gudjonsson JE. Sex bias in autoimmunity. Curr Opin Rheumatol. 2019;31:53–61. Tan K, Yeo T, Yong KP, Thomas T, Wang F, Tye J, Quek AML. Central nervous system inflammatory demyelinating diseases and neuroimmunology in Singapore—epidemiology and evolution of an emerging subspecialty. Neurol Clin Neurosci. 2021:1–7. Caron-Cantin M, Cestari DM, Fortin E. Clinical and radiologic approach to ’typical’ versus antibody-related optic neuritis. Curr Opin Ophthalmol. 2019;30:412–417. Zhou H, Xu Q, Zhao S, Wang W, Wang J, Chen Z, Lin D, Li X, Peng C, Ai N, Wei S. Distinct clinical characteristics of atypical optic neuritis with seronegative aquaporin-4 antibody among Chinese patients. Br J Ophthalmol. 2017;101:1720–1724. Akman-Demir G, Tuzun E, Waters P, Icoz S, Kurtuncu M, Jarius S, Yapici Z, Mutlu M, Yesilot N, Vincent A, Eraksoy M. Prognostic implications of aquaporin-4 antibody status in neuromyelitis optica patients. J Neurol. 2011;258:464–470. Ciotti JR, Eby NS, Wu GF, Naismith RT, Chahin S, Cross AH. Clinical and laboratory features distinguishing MOG antibody disease from multiple sclerosis and AQP4 antibody-positive neuromyelitis optica. Mult Scler Relat Disord. 2020;45:102399. Masuda H, Mori M, Uzawa A, Muto M, Uchida T, Ohtani R, Akiba R, Yokouchi H, Yamamoto S, Kuwabara S. Recovery from optic neuritis attack in neuromyelitis optica spectrum disorder and multiple sclerosis. J Neurol Sci. 2016;367:375–379. Seay M, Rucker JC. Neuromyelitis optica: review and utility of testing aquaporin-4 antibody in typical optic neuritis. Asia Pac J Ophthalmol (Phila). 2018;7:229–234. Mealy MA, Whetstone A, Orman G, Izbudak I, Calabresi PA, Levy M. Longitudinally extensive optic neuritis as an MRI biomarker distinguishes neuromyelitis optica from multiple sclerosis. J Neurol Sci. 2015;355:59–63. Freitas E, Guimarães J. Neuromyelitis optica spectrum disorders associated with other autoimmune diseases. Rheumatol Int. 2015;35:243–253. Carnero Contentti E, De Virgiliis M, Hryb JP, Leguizamon F, Celson J, Di Pace JL, Perassolo M. Diagnostic utility of systematic aquaporin-4 antibodies determination in the first event of immune-mediated optic neuritis. Eur Neurol. 2016;76:227–233. 461 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.