Optic Neuritis-Independent Retinal Atrophy in Neuromyelitis Optica Spectrum Disorder

Background: A limited number of studies have investigated the presence of ongoing disease activity independent of clinical relapses in neuromyelitis optica spectrum disorder (NMOSD), and data are conflicting. The objective of our study was to examine whether patients with aquaporin-4 (AQP4)-IgG seropositive NMOSD exhibit progressive retinal neuroaxonal loss, independently of optic neuritis (ON) attacks. Methods: In this single-center, longitudinal study, 32 AQP4-IgG+ NMOSD patients and 48 healthy controls (HC) were followed with serial spectral-domain optical coherence tomography and visual acuity (VA) assessments. NMOSD patients with ON less than 6 months before baseline were excluded, whereas data from patients with ON during follow-up were censored at the last visit before ON. VA worsening was defined as a decrease in monocular letter acuity ≥5 letters for high-contrast VA and ≥7 letters for low-contrast VA. Analyses were performed with mixed-effects linear regression models adjusted for age, sex, and race. Results: The median follow-up duration was 4.2 years (interquartile range: 1.8-7.5). Relative to HC, NMOSD eyes had faster peripapillary retinal nerve fiber layer (pRNFL) (β = -0.25 µm/year faster, 95% confidence interval [CI]: -0.45 to -0.05, P = 0.014) and GCIPL thinning (β = -0.09 µm/year faster, 95% CI: -0.17 to 0, P = 0.05). This difference seemed to be driven by faster pRNFL and GCIPL thinning in NMOSD eyes without a history of ON compared with HC (GCIPL: β = -0.15 µm/year faster; P = 0.005; pRNFL: β = -0.43 µm/year faster, P < 0.001), whereas rates of pRNFL (β: -0.07 µm/year, P = 0.53) and GCIPL (β = -0.01 µm/year, P = 0.90) thinning did not differ between NMOSD-ON and HC eyes. Nine NMOSD eyes had VA worsening during follow-up. Conclusions: In this longitudinal study, we observed progressive pRNFL and GCIPL atrophy in AQP4-IgG+ NMOSD eyes unaffected by ON. These results support that subclinical involvement of the anterior visual pathway may occur in AQP4-IgG+ NMOSD.

Original Contribution Section Editors: Clare Fraser, MD Susan Mollan, MD Optic Neuritis–Independent Retinal Atrophy in Neuromyelitis Optica Spectrum Disorder Angeliki G. Filippatou, MD, Eleni S. Vasileiou, MD, Yufan He, PhD, Kathryn C. Fitzgerald, ScD, Grigorios Kalaitzidis, MD, Jeffrey Lambe, MBBCh, MRCPI, Maureen A. Mealy, PhD, RN, Michael Levy, MD, PhD, Yihao Liu, MSc, Jerry L. Prince, PhD, Ellen M. Mowry, MD, MCR, Shiv Saidha, MBBCh, MD, MRCPI, Peter A. Calabresi, MD, Elias S. Sotirchos, MD Background: A limited number of studies have investigated the presence of ongoing disease activity independent of clinical relapses in neuromyelitis optica spectrum disorder (NMOSD), and data are conflicting. The objective of our study was to examine whether patients with aquaporin-4 (AQP4)-IgG seropositive NMOSD exhibit progressive retinal neuroaxonal loss, independently of optic neuritis (ON) attacks. Methods: In this single-center, longitudinal study, 32 AQP4IgG+ NMOSD patients and 48 healthy controls (HC) were followed with serial spectral-domain optical coherence tomography and visual acuity (VA) assessments. NMOSD patients with ON less than 6 months before baseline were excluded, whereas data from patients with ON during followup were censored at the last visit before ON. VA worsening was defined as a decrease in monocular letter acuity $5 letters for high-contrast VA and $7 letters for low-contrast VA. Analyses were performed with mixed-effects linear regression models adjusted for age, sex, and race. Results: The median follow-up duration was 4.2 years (interquartile range: 1.8–7.5). Relative to HC, NMOSD eyes had faster peripapillary retinal nerve fiber layer (pRNFL) (b = 20.25 mm/year faster, 95% confidence interval [CI]: 20.45 to 20.05, P = 0.014) and GCIPL thinning (b = 20.09 mm/year faster, 95% CI: 20.17 to 0, P = 0.05). This difference seemed to be driven by faster pRNFL and GCIPL thinning in NMOSD eyes without a history of ON compared with HC (GCIPL: b = 20.15 mm/year faster; P = 0.005; pRNFL: b = 20.43 mm/ year faster, P , 0.001), whereas rates of pRNFL (b: 20.07 mm/year, P = 0.53) and GCIPL (b = 20.01 mm/year, P = 0.90) thinning did not differ between NMOSD-ON and HC eyes. Nine NMOSD eyes had VA worsening during follow-up. Conclusions: In this longitudinal study, we observed progressive pRNFL and GCIPL atrophy in AQP4-IgG+ NMOSD eyes unaffected by ON. These results support that subclinical involvement of the anterior visual pathway may occur in AQP4-IgG+ NMOSD. Journal of Neuro-Ophthalmology 2022;42:e40–e47 doi: 10.1097/WNO.0000000000001282 © 2021 by North American Neuro-Ophthalmology Society Department of Neurology (AGF, ESV, KCF, GK, JL, MAM, EMM, SS, PAC, ESS), Johns Hopkins University School of Medicine, Baltimore, Maryland; Department of Electrical and Computer Engineering (YH, YL, JLP), Johns Hopkins University, Baltimore, Maryland; Viela Bio (MAM), Gaithersburg, Maryland; and Department of Neurology (ML), Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts. Supported by the Caring Friends NMO Research fund, National MS Society (FP-1607-24999 to ESS, RG-1606-08768 to SS; and TA-1805-31136 to KCF), NIH/NINDS (R01NS082347 to PAC and K23NS117883 to ESS), and NIH/NIMH (K01MH121582 to KCF). M. A. Mealy is an employee of Viela Bio. M. Levy has received research support from: National Institutes of Health, Maryland Technology Development Corporation, Sanofi, Genzyme, Alexion, Alnylam, Shire, Acorda, and Apopharma. He has also received personal compensation for consultation with Alexion, Acorda, and Genzyme, and he serves on the scientific advisory boards for Alexion, Acorda, and Quest Diagnostics. J. Prince is a founder of Sonovex, Inc. and serves on its Board of Directors. He has received consulting fees from JuneBrain LLC and is PI on research grants to Johns Hopkins from Biogen. E. Mowry has grants from Biogen and Genzyme, is site PI for studies sponsored by Biogen, has received free medication for a clinical trial from Teva, and receives royalties for editorial duties from UpToDate. S. Saidha has received consulting fees from Medical Logix for the development of CME programs in neurology and has served on scientific advisory boards for Biogen, Genzyme, Genentech Corporation, EMD Serono, and Celgene. He is the PI of investigator-initiated studies funded by Genentech Corporation and Biogen and received support from the Race to Erase MS foundation. He has received equity compensation for consulting from JuneBrain LLC, a retinal imaging device developer. He is also the site investigator of a trial sponsored by MedDay Pharmaceuticals. P. A. Calabresi has received consulting fees from Disarm Therapeutics and Biogen and is PI on grants to JHU from Biogen and Annexon. E. S. Sotirchos has served on scientific advisory boards for Viela Bio, Genentech, and Alexion and has received speaker honoraria from Viela Bio and Biogen. Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Web site (www.jneuro-ophthalmology.com). A. G. Filippatou and E. S. Vasileiou contributed equally to this work. Address correspondence to: Elias S. Sotirchos, Division of Neuroimmunology and Neurological Infections, Department of Neurology, Pathology 627, 600 N. Wolfe Street, Baltimore, MD 21287; E-mail: [e-mail] e40 Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution N euromyelitis optica spectrum disorder (NMOSD) is a rare immune-mediated disease of the central nervous system, associated, in most cases, with autoantibodies directed against the astrocytic water channel aquaporin-4 (AQP4) (1). AQP4-IgG+ NMOSD presents predominantly with episodes of optic neuritis (ON) and longitudinally extensive transverse myelitis. Unlike other neuroinflammatory diseases, such as multiple sclerosis (MS), where a progressive disease course has been well established, AQP4IgG+ NMOSD is considered a primarily relapsing disorder and disease progression independent of attack-associated incidents is considered exceedingly rare (1,2). A limited number of studies have investigated the presence of ongoing disease activity independent of relapses in NMOSD, and data are conflicting. Cross-sectional studies using retinal optical coherence tomography (OCT) have suggested that retinal ganglion cell damage may occur in NMOSD eyes in the absence of a history of ON, as evidenced by lower thickness of the inner retinal layers when compared with healthy controls (HCs) (3–5). To date, few longitudinal studies have explored whether there is subclinical visual pathway involvement outside of ON incidents in NMOSD. Although some studies have suggested that progressive visual evoked potentials (VEP) latency delay and progressive GCIPL atrophy may occur in NMOSD even in the absence of ON attacks, existing evidence is conflicting (6–10). Using OCT, the objective of our study was to examine whether AQP4-IgG+ NMOSD patients exhibit progressive retinal neuro-axonal loss, independently of ON attacks. METHODS Standard Protocol Approvals, Registrations, and Patient Consents The study was approved by the Johns Hopkins University Institutional Review Board. All participants provided written informed consent. Study Design and Participants Patients were recruited from the Johns Hopkins Neuromyelitis Optica Clinic between 2008 and 2020 and fulfilled the 2015 International Panel for Neuromyelitis Optica Diagnosis criteria for AQP4-IgG+ NMOSD (retroactively applied to patients recruited before 2015) (1). AQP4-IgG+ NMOSD patients underwent serial retinal imaging with OCT and visual function assessments at their clinical visits. HCs were recruited by convenience sampling and were invited for annual OCT scans. Patients with a clinical history of ON less than 6 months before baseline were excluded because acute ON events are expected to lead to significant inner retinal atrophy. Existing Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 literature supports that most GCIPL and peripapillary retinal nerve fiber layer (pRNFL) thinning, as well as maximal visual recovery, has already occurred by 6 months (11–13). Data of subjects who developed ON during follow-up were censored at the last available scan before ON. Because chiasmal/optic tract involvement is common in ON in AQP4-IgG+ NMOSD, both eyes were excluded or censored in the analyses, even in cases of apparent unilateral ON during follow-up (1). Only eyes with at least 6 months of OCT follow-up were included. Subjects with history of diabetes mellitus, uncontrolled hypertension, or other significant ophthalmological/ neurological disorders were excluded from the study. Optical Coherence Tomography Retinal imaging was performed with spectral-domain OCT (Cirrus HD-OCT, Model 5000; Carl Zeiss Meditec, Dublin, CA), as previously described (14). In brief, pRNFL thicknesses were derived by the conventional Cirrus HDOCT software. Macular ganglion cell + inner plexiform layer (GCIPL) was automatically segmented and thickness was calculated within an annulus centered at the fovea, with an internal diameter of 1 mm and an external diameter of 5 mm. The scans underwent rigorous quality control in accordance with the OSCAR-IB criteria; scans with signal strength below 7/10, with artifact, or with segmentation errors were excluded from the study (15). Visual Function The monocular visual acuity (VA) was assessed using standardized retroilluminated eye charts (Precision Vision, La Salle, IL) with habitual correction. The high-contrast (100%) letter acuity (HCLA) was assessed with Early Treatment Diabetic Retinopathy Study charts (at 4 m) and low-contrast (2.5% and 1.25%) letter acuity (LCLA) with Sloan Letter charts (at 2 m). The maximum score for each chart is 70 letters, corresponding to a Snellen VA of 20/10. Clinically significant VA worsening was defined as a decrease in monocular letter acuity $5 letters for HCLA, and $7 letters for LCLA, in accordance with prior studies in MS (16,17). Statistical Methods Statistical analyses were performed using Stata version 16 (StataCorp) and R version 4.0.2. Baseline comparisons were performed with mixed-effects linear regression models with subject-specific random intercepts. Rates of change of retinal layer thicknesses and letter acuity scores were analyzed using mixed-effects linear regression models, with subject-specific and eye-specific random intercepts and random slopes in time, using time from baseline visit as a continuous variable. All analyses were adjusted for age, sex, and race. Statistical significance was defined as P , 0.05. e41 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution RESULTS Cohort 32 AQP4-IgG+ NMOSD patients (61 eyes) met inclusion criteria (see Supplemental Digital Content, Figure E1, http://links.lww.com/WNO/A467), and we also included 48 HC (95 eyes), approximately matched to the NMOSD group for age, sex, and race. Demographics and clinical characteristics are summarized in Table 1. The median OCT follow-up was 4.3 years in NMOSD and 4.0 years in HC. For a subset of NMOSD patients (n = 30) and HC (n = 42), VA assessments were available cross-sectionally. In addition, 20 NMOSD patients (63%) had available serial VA assessments (median follow-up 3.0 years and interquartile range: 1.0–7.6). Baseline Optical Coherence Tomography Analysis At baseline, GCIPL and pRNFL thicknesses were markedly lower in NMOSD eyes with a ON history (NMOSD-ON) compared with HCs (adjusted mean differences—GCIPL: 220.29 mm; pRNFL: 227.80 mm; P , 0.001 for both; Table 2). NMOSD eyes without a ON history (NMOSD–nonON) also exhibited significantly lower GCIPL thickness, albeit less pronounced relative to NMOSD-ON eyes, as compared with HCs (adjusted mean difference: 25.38 mm, P , 0.001). pRNFL thickness was also lower in NMOSD–non-ON eyes compared with HCs, but this difference did not attain statistical significance (adjusted mean difference: 24.06 mm, P = 0.09; Table 2). Longitudinal Optical Coherence Tomography Analysis Mean annualized rates of change in GCIPL and pRNFL thicknesses in NMOSD and HC participants are reported in Table 3, and individual eye trajectories are shown in Figure 1. NMOSD patients exhibited faster rates of pRNFL thinning compared with HCs (20.45 mm/year vs 20.21 mm/year, b = 20.25 mm/year, 95% CI: 20.45 to 20.05, P = 0.014). Similarly, GCIPL thinning was also accelerated in NMOSD patients (20.29 mm/year vs 20.20 mm/year, b = 20.09 mm/year, 95% CI: 20.17 to 0.00, P = 0.05). These differences correspond to 119% faster pRNFL and 45% faster GCIPL thinning in NMOSD compared with HC eyes. This was driven by faster pRNFL and GCIPL thinning in NMOSD–non-ON eyes compared with HC (pRNFL: 20.64 mm/year vs 20.21 mm/year, b = 20.43 mm/year, 95% CI: 20.67 to 20.19, P , 0.001; GCIPL: 20.35 mm/year vs 20.20 mm/year, b = 20.15 mm/year, 95% CI: 20.25 to 20.05, P = 0.005). By contrast, NMOSD- TABLE 1. Demographic and clinical characteristics HC Participants, n Age in years, mean (SD) Female, n (%) Race, n (%) Caucasian American African American Asian American Relapse history, n (%) ON & TM ON & TM & brainstem attack Isolated TM Isolated ON Brainstem attack and TM Treatment during follow-up, patient-years (%) No treatment Rituximab Mycophenolate mofetil Azathioprine Eculizumab Eculizumab + mycophenolate mofetil Inebilizumab Eyes with ON history, n (%) Disease duration; median (IQR) Follow-up, median (IQR) AQP4-IgG+ NMOSD 48 40.5 (11.3) 38 (79%) 32 45.5 (12.3) 27 (84%) 29 (60%) 18 (38%) 1 (2%) — 15 (47%) 16 (50%) 1 (3%) P 0.063 0.77 0.50 — 13 (41%) 6 (19%) 8 (25%) 2 (6%) 3 (9%) — — — 4.0 (1.8–7.5) — 11.3 (8.7%) 86.3 (66.7%) 15.9 (12.3%) 4.7 (3.6%) 3.8 (2.9%) 6.0 (4.6%) 1.4 (1.1%) 34 (55%)* 6 (1–9) 4.3 (2.6–7.5) — — 0.89 *8 NMOSD-ON eyes had microcystic macular pathology (MMP). HC, healthy controls; AQP4-IgG+ NMOSD, aquaporin-4-IgG seropositive neuromyelitis optica spectrum disorder; TM, transverse myelitis; ON, optic neuritis; IQR, interquartile range. e42 Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution TABLE 2. Cross-sectional OCT measures and visual acuity scores in AQP4-IgG+ NMOSD and HC eyes and comparisons between groups HC (n = 95 Eyes) NMOSD–non-ON (n = 30 Eyes) NMOSD-ON (n = 31 Eyes) 76.9 (5.2) 93.9 (9.2) 71.6 (4.8) 91.2 (8.2) 56.7 (8.2) 67.3 (13.1) 61.5 (59–65) 34 (29–38) 20 (14–25) 59 (51.5–63) 24 (16–33) 6.5 (5–18) 44.5 (7–54) 0 (0–16.5) 0 (0–1.5) OCT Measures, mm, Mean (SD) GCIPL pRNFL Letter acuity scores†, median (IQR) 100% contrast (HCLA) 2.5% contrast (LCLA) 1.25% contrast (LCLA) HC vs NMOSD–non-ON HC vs NMOSD-ON Difference (95% CI)* OCT Measures, mm GCIPL pRNFL Letter acuity scores† 100% contrast (HCLA) 2.5% contrast (LCLA) 1.25% contrast (LCLA) P Difference (95% CI)* P 25.38 (28.16 to 22.60) 24.06 (28.78 to 0.66) ,0.001 0.09 220.29 (223.37 to 217.20) 227.80 (232.62 to 222.98) ,0.001 ,0.001 22.42 (25.24 to 0.40) 28.15 (212.15 to 24.15) 28.66 (213.24 to 24.08) 0.10 ,0.001 ,0.001 223.00 (228.92 to 217.07) 223.95 (227.93 to 219.96) 216.97 (220.90 to 213.05) ,0.001 ,0.001 ,0.001 Statistically significant results are in bold. *Derived from mixed-effects linear regression models, including age, sex, and race. † Available for 42 HC and 30 NMOSD patients. OCT, optical coherence tomography; AQP4, aquaporin-4; NMOSD, neuromyelitis optica spectrum disorder; HC, healthy controls; ON, optic neuritis; CI, confidence interval; GCIPL, ganglion cell + inner plexiform layer; pRNFL, peripapillary retinal nerve fiber layer; IQR, interquartile range; HCLA, high-contrast letter acuity; LCLA, low-contrast letter acuity. tial floor effect. The rate of GCIPL thinning was also faster in NMOSD-ON eyes with baseline pRNFL thickness $70 mm compared with HCs, although this finding was not statistically significant (20.21 mm/year vs 20.32 mm/ year, b = 20.11 mm/year, 95% CI: 20.27 to 0.05, P = 0.16). To exclude effects because of chiasmal or optic tract involvement in patients with a history of contralateral ON, we compared rates of retinal atrophy between NMOSD patients who had never experienced ON in either eye (n = 11 patients) and HCs, and the results were similar (pRNFL: 20.62 mm/year vs 20.21 mm/year, b = 20.41 ON eyes did not demonstrate significant differences in rates of pRNFL or GCIPL thinning in comparison with HCs (pRNFL: 20.28 mm/year vs 20.21 mm/year, b = 20.07 mm/year, 95% CI: 20.31 to 0.16, P = 0.53; GCIPL: 20.21 mm/year vs 20.20 mm/year, b = 20.01 mm/year, 95% CI: 20.11 to 0.10, P = 0.90). However, when restricting analyses to NMOSD-ON eyes with baseline pRNFL thickness $70 mm (lower limit of pRNFL thickness range in the HC cohort; n = 12 NMOSD-ON eyes), we found faster rates of pRNFL thinning as compared with HCs (20.57 vs 20.21 mm/year, b = 20.36 mm/year, 95% CI: 20.57 to 20.21, P = 0.034), consistent with a poten- TABLE 3. Annualized change in retinal layer thickness in AQP4-IgG+ NMOSD and HC eyes and comparisons between groups Annualized Rate of Change, mm/year HC (n = 95 Eyes) NMOSD (n = 61 Eyes) HC vs NMOSD NMOSD–non-ON (n = 30 Eyes) NMOSD-ON (n = 31 Eyes) GCIPL 20.20 (20.25 to 20.15)20.29 (20.36 to 20.22)20.35 (20.44 to 20.26)20.21 (20.30 to 20.12) pRNFL20.21 (20.33 to 20.08)20.45 (20.61 to 20.30)20.64 (20.86 to 20.43)20.28 (20.48 to 20.07) HC vs NMOSD–non-ON GCIPL pRNFL Difference in Annualized Rate of Change (95% CI),* mm/year P 20.09 (20.17 to 0.00) 20.25 (20.45 to 20.05) 0.05 0.014 HC vs NMOSD-ON Difference in Annualized Rate of Change (95% CI),* mm/year P Difference in Annualized Rate of Change (95% CI),* mm/year P 20.15 (20.25 to 20.05) 20.43 (20.67 to 20.19) 0.005 ,0.001 20.01 (20.11 to 0.10) 20.07 (20.31 to 0.16) 0.90 0.53 Statistically significant results are in bold. *Derived from mixed-effects linear regression models, including age, sex, and race. AQP4, aquaporin-4; NMOSD, neuromyelitis optica spectrum disorder; HC, healthy controls; ON, optic neuritis; CI, confidence interval; GCIPL, ganglion cell + inner plexiform layer; pRNFL, peripapillary retinal nerve fiber layer. Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 e43 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution FIG. 1. Longitudinal changes in pRNFL and GCIPL thickness. Longitudinal trajectories of GCIPL and pRNFL thickness for individual eyes are shown, separately for healthy controls, AQP4-IgG+ NMOSD eyes without a history of ON (NMOSD–non-ON) and with a history of ON (NMOSD-ON). The thicker lines represent the average trajectory, as fitted by mixed-effects models. The y-axis range differs for the NMOSD-ON group (given lower GCIPL and pRNFL thicknesses), but the slopes are comparable across panels because the scales of the y-axes are identical. NMOSD: neuromyelitis optica spectrum disorder; ON: optic neuritis; GCIPL: ganglion cell + inner plexiform layer; pRNFL: peripapillary retinal nerve fiber layer. mm/year, 95% CI: 20.68 to 20.15, P = 0.002; GCIPL: 20.31 mm/year vs 20.20 mm/year, b = 20.11 mm/year, 95% CI: 20.22 to 0.00, P = 0.05). Finally, we explored the effects of non-ON relapses during follow-up on rates of pRNFL and GCIPL thinning. Ten patients had relapses during follow-up (9 transverse myelitis and 1 area postrema syndrome). Patients with relapses did not exhibit significant differences in rates of GCIPL (20.27 mm/ year vs 20.32 mm/year, b = 0.05 mm/year, 95% CI: 20.10 to 0.20, P = 0.51) or pRNFL thinning (20.44 mm/year vs 20.51 mm/year, b = 0.08 mm/year, 95% CI: 20.28 to 0.43; P = 0.67), compared with those who were clinically stable. Baseline Visual Acuity Analysis At baseline, as expected, HCLA and LCLA were markedly worse in NMOSD-ON eyes compared with HCs (adjusted mean differences—HCLA: 223.00 letters; 2.5% LCLA: 223.95 letters; 1.25% LCLA: 216.97 letters, P , 0.001 for all; Table 2). HCLA did not differ between NMOSD–non-ON and HC eyes (adjusted mean difference: 22.42 letters, P = 0.10); however, NMOSD–non-ON eyes exhibited e44 worse 2.5% and 1.25% LCLA compared with HCs (adjusted mean differences: 2.5% LCLA: 28.15 letters; 1.25% LCLA: 28.66 letters, P , 0.001 for both; Table 2). Longitudinal Visual Acuity Analysis Of the 61 NMOSD eyes included in this study, 40 eyes had serial visual function assessments. Of these 40 eyes, 6 eyes (15%) exhibited clinically significant HCLA worsening by the end of the follow-up period, 4 eyes (10%) exhibited clinically significant 2.5% LCLA worsening, and 5 eyes (13%) exhibited clinically significant 1.25% LCLA worsening. Nine eyes (23%) exhibited HCLA and/or LCLA worsening. Rates of GCIPL and pRNFL thinning were compared between eyes with clinically significant VA worsening by the end of the follow-up period and eyes with stable VA. We did not find significant differences in these rates between eyes with VA worsening compared with those with stable VA (GCIPL: 20.28 mm/year vs 20.33 mm/year, b = 0.05 mm/year, 95% CI: 20.11 to 0.22, P = 0.51; pRNFL: 20.63 mm/year vs 20.51 mm/year, b = 20.12 mm/year, 95% CI: 20.55 to 0.32, P = 0.6). Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution We also calculated the annualized rates of change of HCLA and LCLA eyes in NMOSD eyes and did not observe a significant change in visual function over time (Table 4). There was no difference in the rate of change of HCLA and LCLA between NMOSD-ON and NMOSD– non-ON eyes. DISCUSSION In this longitudinal study, we observed accelerated pRNFL and GCIPL thinning in AQP4-IgG+ NMOSD eyes clinically unaffected by ON. This finding suggests that subclinical progressive retinal neuroaxonal loss occurs in AQP4IgG+ NMOSD. This has important implications for our understanding of the disease mechanisms of NMOSD because it provides evidence that there is ongoing insidious disease activity independent of clinical relapses. Importantly, our results remained unaltered in sensitivity analyses that were restricted to patients who had never experienced ON in either eye, suggesting that our findings are not the consequence of a symptomatic ON event but they likely reflect a separate pathophysiologic process. Our findings are in agreement with prior work by Oertel et al reporting faster GCIPL thinning in AQP4-IgG+ NMOSD eyes, independently of ON attacks (6). Furthermore, in a multicenter longitudinal VEP study, an increase of P100 latencies and a decline of amplitudes was observed in AQP4-IgG+ NMOSD eyes without ON during followup (7). Together, these findings support that the visual pathway may be a site of subclinical involvement in AQP4-IgG+ NMOSD. In our study, these phenomena seemed to be independent, not only of ON attacks, but also of other non-ON attacks, because pRNFL and GCIPL atrophy rates did not differ between patients who experienced relapses during follow-up and patients who were clinically stable. In a cohort of 27 NMOSD patients, Pisa et al reported accelerated pRNFL thinning in patients with dis- ease activity during follow-up, but not in stable patients (18). By contrast, in the study by Oertel et al, patients who experienced attacks during follow-up had pRNFL thickening (6). Further larger studies are likely needed to investigate the retinal phenomena that occur during inflammatory attacks and reconcile these conflicting results. Interestingly, we observed progressive pRNFL and GCIPL atrophy in AQP4-IgG+ NMOSD eyes without a history of ON, but not in ON eyes. This is presumably because of a floor effect, because ON associated with AQP4-IgG+ NMOSD results in marked retinal ganglion cell and axonal loss which may limit detection of further decreases in pRNFL or GCIPL thicknesses (19,20). Notably, subgroup analyses restricted to NMOSD-ON eyes with baseline pRNFL thickness $70 mm revealed faster rates of pRNFL thinning compared with HCs, consistent with this hypothesis. Although the pathoetiology of the observed progressive retinal neuroaxonal loss could not be established in the current study, several hypotheses need to be considered. AQP4 is a water channel protein that is highly expressed in the retina by astrocytes and more abundantly by a unique type of glial cell, Müller cells. Müller cells regulate vital retinal homeostatic mechanisms, including release of neurotropic factors and degradation of glutamate. It is therefore conceivable that the observed retinal changes are a consequence of direct retinal damage because of anti-AQP4 antibody activity (21). Interestingly, it has been previously reported that Müller cells are richly expressed in the foveal region, located in the center of the macula, and foveal thinning has been reported in AQP4-IgG+ eyes without a history of ON; it has been postulated that this may represent the occurrence of a primary retinal astrocytopathy (21,22). In a pathological retinal study, AQP4-IgG+ NMOSD was characterized by loss of AQP4 immunoreactivity on Müller cells; it is likely that alterations in the dynamics of astrocyte and Müller cell function may be responsible for subsequent TABLE 4. Annualized change in the visual acuity in AQP4-IgG+ NMOSD eyes Annualized Rate of Change, Letters/year 100% contrast (HCLA) 2.5% contrast (LCLA) 1.25% contrast (LCLA) NMOSD (n = 40 Eyes) NMOSD-nonON (n = 20 Eyes) NMOSD-ON (n = 20 Eyes) 20.04 (20.38 to 0.29) 0.08 (20.39 to 0.55) 0.02 (20.31 to 0.34) 20.10 (20.54 to 0.33) 20.12 (20.71 to 0.48) 0.12 (20.30 to 0.54) 0.03 (20.51 to 0.57) 0.20 (20.51 to 0.91) 20.08 (20.59 to 0.42) NMOSD–non-ON vs NMOSD-ON Difference in Annualized Rate of Change (95% CI)*, Letters/year 0.13 (20.56 to 0.83) 0.31 (20.62 to 1.24) 20.20 (20.86 to 0.45) P 0.71 0.51 0.55 *Derived from mixed-effects linear regression models, including age, sex, and race. AQP4, aquaporin-4; NMOSD, neuromyelitis optica spectrum disorder; ON, optic neuritis; CI, confidence interval; HCLA, high-contrast letter acuity; LCLA, low-contrast letter acuity. Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 e45 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution neurodegeneration (23). Alternatively, our findings may be the result of subclinical involvement of the anterior or posterior visual pathway. A study by Hokari et al reported pathological evidence of optic nerve involvement in AQP4-IgG+ NMOSD eyes with no clinical ON attacks (23). Moreover, gadolinium enhancement of the optic nerve has been reported at the time of attacks involving other anatomical locations, in the absence of visual symptoms (24). Retrograde trans-synaptic degeneration because of involvement of the posterior visual pathway is another consideration. This is a less likely explanation, however, because optic radiation (OR) lesions are not as common in NMOSD relative to MS (25). Moreover, development of new asymptomatic brain lesions in clinically stable AQP4-IgG+ NMOSD patients seems to be rare (26). Furthermore, we did not observe any cases of homonymous GCIPL thinning (which would be expected to be observed in unilateral posterior visual pathway involvement), although this may be difficult to assess in the presence of pre-existing optic neuropathy and/or bilateral posterior visual pathway involvement. Our VA findings are also of interest, because we found that NMOSD–non-ON eyes had worse performance in LCLA charts compared with HC. This suggests that some level of vision impairment may be present in AQP4-IgG+ NMOSD eyes, even without a clinical history of ON, and although the etiology of this finding is not clear, considerations include prior subclinical ON and/or primary retinal astrocytopathy, as outlined above. We also found that 23% of NMOSD eyes had worsening visual function during follow-up, based on cut-offs that have been previously proposed for MS eyes (16). However, rates of retinal thinning did not differ between eyes with VA worsening compared with those with stable VA, and on a group level, we did not observe a significant change in visual function over time. Although this finding may suggest that there is no direct correlation between OCT measures and functional outcomes, this negative finding should be interpreted with caution, given our relatively small sample size and the fact that VA assessments were available for only a subset of our cohort. Alternatively, this result may suggest that OCT is a more sensitive indicator of anterior visual pathway involvement, and retinal neuroaxonal loss may precede visual deficits. One limitation of our study is the relatively the small sample size, which is expected given the rarity of the studied condition. Furthermore, longitudinal VA data were not available for most of the HC participants. It is likely that our study was underpowered to detect potential associations between neuroaxonal loss and vision loss, because of insufficient follow-up time and/or sample size. Moreover, our study was retrospective, and brain and optic nerve MRI scans were not routinely performed in all patients. Therefore, we could not assess for the presence of lesions along the visual pathway. Finally, most participants were on e46 treatment with rituximab during follow-up, and we were therefore unable to investigate any potential associations between different disease-modifying treatments and OCT measures. To conclude, our findings support that subclinical progressive retinal axonal loss may occur in AQP4IgG+ NMOSD. Further studies are warranted to validate these findings and to elucidate their clinical relevance, because the presence of insidious pathology of the visual system has important implications for our understanding of the disease mechanisms of NMOSD and for monitoring treatment response. REFERENCES 1. Wingerchuk DM, Banwell B, Bennett JL, Cabre P, Carroll W, Chitnis T, de Seze J, Fujihara K, Greenberg B, Jacob A, Jarius S, Lana-Peixoto M, Levy M, Simon JH, Tenembaum S, Traboulsee AL, Waters P, Wellik KE, Weinshenker BG. International consensus diagnostic criteria for neuromyelitis optica spectrum disorders. Neurology. 2015;85:177–189. 2. Wingerchuk DM, Pittock SJ, Lucchinetti CF, Lennon VA, Weinshenker BG. A secondary progressive clinical course is uncommon in neuromyelitis optica. Neurology. 2007;68:603– 605. 3. Tian D, Su L, Fan M, Yang J, Zhang R, Wen P, Han Y, Yu C, Zhang C, Ren H, Shi K, Zhu Z, Dong Y, Liu Y, Shi F. Bidirectional degeneration in the visual pathway in neuromyelitis optica spectrum disorder (NMOSD). Mult Scler. 2018;24:1585– 1593. 4. Sotirchos ES, Saidha S, Byraiah G, Mealy MA, Ibrahim MA, Sepah YJ, Newsome SD, Ratchford JN, Frohman EM, Balcer LJ, Crainiceanu CM, Nguyen QD, Levy M, Calabresi PA. In vivo identification of morphologic retinal abnormalities in neuromyelitis optica. Neurology. 2013;80:1406–1414. 5. Filippatou AG, Vasileiou ES, He Y, Fitzgerald KC, Kalaitzidis G, Lambe J, Mealy MA, Levy M, Liu Y, Prince JL, Mowry EM, Saidha S, Calabresi PA, Sotirchos ES. Evidence of subclinical quantitative retinal layer abnormalities in AQP4-IgG seropositive NMOSD. Mult Scler. 2020:1352458520977771. 6. Oertel FC, Havla J, Roca-Fernández A, Lizak N, Zimmermann H, Motamedi S, Borisow N, White OB, Bellmann-Strobl J, Albrecht P, Ruprecht K, Jarius S, Palace J, Leite MI, Kuempfel T, Paul F, Brandt AU. Retinal ganglion cell loss in neuromyelitis optica: a longitudinal study. J Neurol Neurosurg Psychiatry. 2018;89:1259–1265. 7. Ringelstein M, Harmel J, Zimmermann H, Brandt AU, Paul F, Haarmann A, Buttmann M, Hümmert MW, Trebst C, Schroeder C, Ayzenberg I, Kleiter I, Hellwig K, Havla J, Kümpfel T, Jarius S, Wildemann B, Rommer P, Weber MS, Pellkofer H, Röpke L, Geis C, Retzlaff N, Zettl U, Deppe M, Klotz L, Young K, Stellmann J, Kaste M, Kermer P, Marouf W, Lauda F, Tumani H, Graf J, Klistorner A, Hartung H, Aktas O, Albrecht P. Longitudinal optic neuritis-unrelated visual evoked potential changes in NMO spectrum disorders. Neurology. 2020;94:e407–e418. 8. Oertel FC, Zimmermann H, Paul F, Brandt AU. Optical coherence tomography in neuromyelitis optica spectrum disorders: potential advantages for individualized monitoring of progression and therapy. EPMA J. 2018;9:21–33. 9. Bouyon M, Collongues N, Zéphir H, Ballonzoli L, Jeanjean L, Lebrun C, Chanson J, Blanc F, Fleury M, Outteryck O, Defoort S, Labauge P, Vermersch P, Speeg C, De Seze J. Longitudinal follow-up of vision in a neuromyelitis optica cohort. Mult Scler. 2013;19:1320–1322. 10. Manogaran P, Traboulsee AL, Lange AP. Longitudinal study of retinal nerve fiber layer thickness and macular volume in patients with neuromyelitis optica spectrum disorder. J Neuroophthalmol. 2016;36:363–368. Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited. Original Contribution 11. Kupersmith MJ, Garvin MK, Wang J, Durbin M, Kardon R. Retinal ganglion cell layer thinning within one month of presentation for optic neuritis. Mult Scler. 2016;22:641–648. 12. Syc SB, Saidha S, Newsome SD, Ratchford JN, Levy M, Ford E, Crainiceanu CM, Durbin MK, Oakley JD, Meyer SA, Frohman EM, Calabresi PA. Optical coherence tomography segmentation reveals ganglion cell layer pathology after optic neuritis. Brain. 2012;135:521–533. 13. Andorrà M, Alba-Arbalat S, Camos-Carreras A, Gabilondo I, Fraga-Pumar E, Torres-Torres R, Pulido-Valdeolivas I, TerceroUribe AI, Guerrero-Zamora AM, Ortiz-Perez S, Zubizarreta I, Sola-Valls N, Llufriu S, Sepulveda M, Martinez-Hernandez E, Armangue T, Blanco Y, Villoslada P, Sanchez-Dalmau B, Saiz A, Martinez-Lapiscina EH. Using acute optic neuritis trials to assess neuroprotective and remyelinating therapies in multiple sclerosis. JAMA Neurol. 2020;77:234–244. 14. 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. 2019:1352458519864928. 15. Tewarie P, Balk L, Costello F, Green A, Martin R, Schippling S, Petzold A. The OSCAR-IB consensus criteria for retinal OCT quality assessment. PLoS One. 2012;7:e34823. 16. Talman LS, Bisker ER, Sackel DJ, Long DA, Galetta KM, Ratchford JN, Lile DJ, Farrell SK, Loguidice MJ, Remington G, Conger A, Frohman TC, Jacobs DA, Markowitz CE, Cutter GR, Ying G, Dai Y, Maguire MG, Galetta SL, Frohman EM, Calabresi PA, Balcer LJ. Longitudinal study of vision and retinal nerve fiber layer thickness in multiple sclerosis. Ann Neurol. 2010;67:749–760. 17. Balcer LJ, Baier ML, Pelak VS, Fox RJ, Shuwairi S, Galetta SL, Cutter GR, Maguire MG. New low-contrast vision charts: reliability and test characteristics in patients with multiple sclerosis. Mult Scler. 2000;6:163–171. 18. Pisa M, Ratti F, Vabanesi M, Radaelli M, Guerrieri S, Moiola L, Martinelli V, Comi G, Leocani L. Subclinical neurodegeneration Filippatou, Vasileiou et al: J Neuro-Ophthalmol 2022; 42: e40-e47 19. 20. 21. 22. 23. 24. 25. 26. in multiple sclerosis and neuromyelitis optica spectrum disorder revealed by optical coherence tomography. Mult Scler. 2019:1352458519861603. Costello F. Optical coherence tomography in neuroophthalmology. Neurol Clin. 2017;35:153–163. Filippatou AG, Mukharesh L, Saidha S, Calabresi PA, Sotirchos ES. AQP4-IgG and MOG-IgG related optic neuritis-prevalence, optical coherence tomography findings, and visual outcomes: a systematic Review and meta-analysis. Front Neurol. 2020;11:540156. You Y, Zhu L, Zhang T, Shen T, Fontes A, Yiannikas C, Parratt J, Barton J, Schulz A, Gupta V, Barnett MH, Fraser CL, Gillies M, Graham SL, Klistorner A. Evidence of müller glial dysfunction in patients with aquaporin-4 immunoglobulin G-positive neuromyelitis optica spectrum disorder. Ophthalmology. 2019;126:801–810. Oertel FC, Kuchling J, Zimmermann H, Chien C, Schmidt F, Knier B, Bellmann-Strobl J, Korn T, Scheel M, Klistorner A, Ruprecht K, Paul F, Brandt AU. Microstructural visual system changes in AQP4-antibody-seropositive NMOSD. Neurol Neuroimmunol Neuroinflamm. 2017;4:e334. Hokari M, Yokoseki A, Arakawa M, Saji E, Yanagawa K, Yanagimura F, Toyoshima Y, Okamoto K, Ueki S, Hatase T, Ohashi R, Fukuchi T, Akazawa K, Yamada M, Kakita A, Takahashi H, Nishizawa M, Kawachi I. Clinicopathological features in anterior visual pathway in neuromyelitis optica. Ann Neurol. 2016;79:605–624. Aksoy D, Gokce E, Kurt S, Cevik B, Demir HD. Subclinical optic neuritis in neuromyelitis optica. J Neuro Ophthalmol. 2013;33:205–207. Shen T, You Y, Arunachalam S, Fontes A, Liu S, Gupta V, Parratt J, Wang C, Barnett M, Barton J, Chitranshi N, Zhu L, Fraser CL, Graham SL, Klistorner A, Yiannikas C. Differing structural and functional patterns of optic nerve damage in multiple sclerosis and neuromyelitis optica spectrum disorder. Ophthalmology. 2019;126:445–453. Lee MY, Yong KP, Hyun J, Kim S, Lee S, Kim HJ. Incidence of inter-attack asymptomatic brain lesions in NMO spectrum disorder. Neurology. 2020;95:e3124–e3128. e47 Copyright © North American Neuro-Ophthalmology Society. Unauthorized reproduction of this article is prohibited.