Neuro-­i nflammation Original research Astrocytic outer retinal layer thinning is not a feature in AQP4-­IgG seropositive neuromyelitis optica spectrum disorders Angelo Lu ‍ ‍,1,2 Hanna G Zimmermann,1,2 Svenja Specovius,1,2 Seyedamirhosein Motamedi,1,2 Claudia Chien ‍ ‍,1,2 Charlotte Bereuter,1,2 Marco A Lana-­Peixoto,3 Mariana Andrade Fontenelle,3 Fereshteh Ashtari,4 Rahele Kafieh,5 Alireza Dehghani,6 Mohsen Pourazizi,6 Lekha Pandit,7 Anitha D’Cunha,7 Ho Jin Kim ‍ ‍,8 Jae-­Won Hyun,8 Su-­Kyung Jung,9 Letizia Leocani,10 Marco Pisa,10 Marta Radaelli,10 Sasitorn Siritho,11 Eugene F May,12 Caryl Tongco,12 Jérôme De Sèze,13 Thomas Senger,13 Jacqueline Palace,14 Adriana Roca-­Fernández ‍ ‍,14 Maria Isabel Leite,14 Srilakshmi M Sharma,15 Hadas Stiebel-­Kalish ‍ ‍,16,17 Nasrin Asgari,18 Kerstin Kathrine Soelberg,19 Elena H Martinez-­Lapiscina ‍ ‍,20 Joachim Havla ‍ ‍,21 Yang Mao-­Draayer,22 Zoe Rimler,23 Allyson Reid,23 Romain Marignier,24 Alvaro Cobo-­Calvo,24,25 Ayse Altintas,26 Uygur Tanriverdi,27 Rengin Yildirim,28 Orhan Aktas,29 ► Additional supplemental Marius Ringelstein ‍ ‍,29,30 Philipp Albrecht ‍ ‍,29 Ivan Maynart Tavares,31 material is published online only. To view, please visit Denis Bernardi Bichuetti ‍ ‍,32 Anu Jacob,33 Saif Huda,33 Ibis Soto de Castillo,34 the journal online (http://​dx.​ Axel Petzold ‍ ‍,35 Ari J Green,36 Michael R Yeaman,37,38 Terry J Smith,39,40 doi.​org/​10.​1136/​jnnp-​2021-​ 327412). Lawrence Cook,41 Friedemann Paul,1,2,42 Alexander U Brandt,1,2,43 For numbered affiliations see Frederike Cosima Oertel ‍ ‍,1,2,36 GJCF International Clinical Consortium for NMOSD end of article. ABSTRACT p=0.027) and MOG-­IgG + (19.82±4.78 µm, p=0.004) Correspondence to Professor Friedemann Paul, Background Patients with anti-­aquaporin-­4 antibody with a history of ON showed parafoveal OPL thinning Experimental and Clinical seropositive (AQP4-­IgG+) neuromyelitis optica spectrum compared with healthy controls (20.99±5.14 µm); this Research Center, Charité disorders (NMOSDs) frequently suffer from optic neuritis was not observed elsewhere. Universitätsmedizin Berlin, (ON) leading to severe retinal neuroaxonal damage. Conclusion The results suggest that outer retinal layer 10117 Berlin, Germany; ​ loss is not a consistent component of retinal astrocytic friedemann.​paul@​charite.​de Further, the relationship of this retinal damage to a primary astrocytopathy in NMOSD is uncertain. Primary damage in AQP4-­IgG+ NMOSD. Longitudinal studies are AB and FCO contributed astrocytopathy has been suggested to cause ON-­ necessary to determine if OPL and ONL are damaged equally. independent retinal damage and contribute to changes in late disease due to retrograde trans-­synaptic axonal particularly in the outer plexiform layer (OPL) and outer degeneration and whether outer retinal dysfunction Received 24 June 2021 Accepted 26 September 2021 nuclear layer (ONL), as reported in some earlier studies. occurs despite any measurable structural correlates. Published Online First 28 However, these were limited in their sample size and October 2021 contradictory as to the localisation. This study assesses outer retinal layer changes using optical coherence INTRODUCTION tomography (OCT) in a multicentre cross-­sectional Neuromyelitis optica spectrum disorders cohort. (NMOSDs) are relapsing autoimmune disor- ► http://​​dx.​​doi.​​org/​​10.​​1136/​ Method 197 patients who were AQP4-­IgG+ and ders affecting the central nervous system (CNS).1 jnnp-​2021-​327846 32 myelin-­oligodendrocyte-­glycoprotein antibody Common clinical attacks in NMOSD include optic seropositive (MOG-­IgG+) patients were enrolled in neuritis (ON), acute myelitis and area postrema this study along with 75 healthy controls. Participants syndrome.2 Serum autoantibodies to aquaporin-­4 © Author(s) (or their underwent neurological examination and OCT with (AQP4-­IgG) are detectable in 60%–80% of patients employer(s)) 2022. Re-­use central postprocessing conducted at a single site. permitted under CC BY-­NC. No with NMOSD .3 4 commercial re-­use. See rights Results No significant thinning of OPL (25.02±2.03 AQP4 is an astrocytic water channel in the CNS.5 and permissions. Published µm) or ONL (61.63±7.04 µm) were observed in patients In the retina, astrocytes are mainly located in the by BMJ. who were AQP4-­IgG+ compared with patients who inner neuroaxonal layers of the retina, but AQP4 To cite: Lu A, Zimmermann were MOG-­IgG + with comparable neuroaxonal damage is additionally highly expressed in retinal Müller HG, Specovius S, et al. J (OPL: 25.10±2.00 µm; ONL: 64.71±7.87 µm) or healthy cells.6 These glial cells have diverse functions, such Neurol Neurosurg Psychiatry controls (OPL: 24.58±1.64 µm; ONL: 63.59±5.78 µm). as regulation of water homeostasis and neurotrans- 2022;93:188–195. Eyes of patients who were AQP4-­IgG+ (19.84±5.09 µm, mitter recycling, and are located around the fovea 188 Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 Neuro-­i nflammation on the exact layers in which these changes occur.8 11 It thereby remains unclear if ORLs, especially the ONL are also potentially affected by primary retinal astrocytopathy in AQP4-­IgG sero- positive NMOSD. Representing the largest international NMOSD dataset collected so far, the CROCTINO study (Collaborative Retro- spective Study on retinal optical coherence tomography (OCT) in Neuromyelitis Optica) overcomes one of the common weak- nesses of NMOSD studies—being limited to small and homog- enous sample populations.14 15 Using OCT data from over 20 centres worldwide, reliable quantitative and qualitative retinal assessment becomes possible, and controversial questions such as ORL changes in AQP4-­IgG seropositive NMOSD can be clar- ified. Apart from patients who were AQP4-­IgG seropositive, the CROCTINO cohort also includes patients with antibodies to myelin-­oligodendrocyte-­glycoprotein (MOG-­IgG); a group that is now believed to be a distinct disease entity.14 16–18 While clini- cally similar and undergoing comparable retinal neurodegenera- tion after ON, MOG-­IgG-­associated disease (MOGAD) lacks an identifiable astrocytopathy component and is thereby an appro- priate diseased control group for patients who were AQP4-­IgG seropositive when investigating astrocytic changes.10 19 In this study, we investigated if ORL thinning, specifically in the foveal and macular ONL, occurs in patients who were AQP4-­IgG seropositive compared with healthy controls (HCs) and with patients with MOGAD as a diseased control group. METHODS Cohort design A total of 539 patients with NMOSD were recruited between 2000 and 2018 as part of CROCTINO (stratified data of centres by device type and number of patients are summarised in the online supplemental file 1).14 Patients with (1) diseases poten- tially confounding OCT analyses (including glaucoma, diabetic retinopathy, retinal surgery and ametropia greater than ±6 diopters), (2) a history of ON within the last 6 months before baseline, (3) no evidence of seropositivity for AQP4-­ IgG or Figure 1 Cohort design and exclusion criteria: from the original 539 MOG-­IgG20 21 and (4) no macular OCT data were excluded. patients recruited in the CROCTINO cohort, 108 patients were excluded Cell-­based assays were used for the detection of AQP4-­IgG and due to missing macular data. Of the remaining 431 patients in the MOG-­IgG antibodies in serum samples from all patients. Clin- segmentation cohort, a further 40 patients were excluded due to anomalies ical data (antibody serology, disease duration, frequency of ON, in their OCT scans (OSCAR-­IB criteria; primarily due to low image location of ON, date of ON, Expanded Disability Standard Scale quality (26 patients) or the presence of microcysts (3 patients) or other and treatment received) were collected from all patients. We also pathologies) or due to data corruption (11 patients). We also excluded included 75 HCs (recruited from Barcelona, Isfahan, Mangalore patients with unknown antibody status (90 patients). Of the remaining and Berlin), who were neither age nor sex matched to either 301 patients, the cohort was split based on AQP4-­IgG or MOG-­IgG cohort. seropositivity and a further set of exclusion criteria were applied based on age (being ≥65 years), ophthalmological comorbidities (eg, glaucoma) and Optical coherence Tomography in instances where follow-­ups occurred within 6 months of an ON attack. Retinal examinations were conducted at each centre using the AQP4-­IgG, anti-­aquaporin-­4 antibody; HC, healthy control; MOG-­IgG, following OCT devices: Spectralis SD-­OCT, Heidelberg Engi- anti-myelin-­oligodendrocyte-­glycoprotein antibody; OCT, optical coherence neering, Heidelberg, Germany (Spectralis), Cirrus HD-­ OCT, tomography; ON, optic neuritis. Carl Zeiss Meditec Inc, Dublin, California, USA (Cirrus) and Topcon 3D-­OCT, Topcon Corp, Tokyo, Japan (Topcon). With respect to each device and each centre, two scans were collected: spanning the entire thickness of the retina.7 Of particular interest (1) a 3.4 mm diameter peripapillary ring scan around the optic is also the Henle Fibre outer nuclear layer (ONL) boundary of nerve head for Spectralis SD-­ OCT (for Cirrus and Topcon the parafovea where AQP4 channels are highly expressed.8 devices: extracted from optic disc volume scans), and (2) a A primary and attack-­independent astrocytopathy in NMOSD macular volume scans, centred on the fovea.14 Scans were cate- has been suggested to contribute to retinal neurodegeneration gorised and uploaded onto a central server to be accessed for and to Müller cell-­ associated parafoveal changes.9–13 Recent further processing. studies suggested potential astrocytopathy-­related outer retinal All OCT images fulfilled the OSCAR-­ IB criteria22 23 (see layer (ORL) thinning in AQP4-­IgG seropositive NMOSD but figure 1—images from 29 patients not fulfilling these criteria were limited in their sample size and in parts contradictory were excluded) and results were presented in line with the Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 189 Neuro-­i nflammation layers were segmented in the following layers: macular retinal Table 1 Demographic overview nerve fibre layer (mRNFL), ganglion cell and inner plexiform HC AQP4-­IgG MOG-­IgG layer (GCIP), inner nuclear layer (INL), outer plexiform layer Subjects 75 197 32 (OPL), ONL, the outer plexiform and nuclear layer (OPNL), (N) photoreceptor layer (PR, inner photoreceptor segments to Number of eyes 148 317 55 (N) Bruch’s membrane) and the total retinal thickness (RT, calcu- Age 32.3±9.6 41.8±12.1 36.5±13.7 lated as the thickness consisting of the RNFL (defined as layer (years, mean±SD) no. 3 per Staurenghi et al26) to the Bruch’s membrane (layer Sex 25 (33.8) 24 (12.2) 10 (31.2) no. 14). All scans were checked and, where necessary, manual (male, N (%)) correction of the automatic segmentation was conducted using EDSS – 3.5 (2.0–5.0) 2.0 (1.5–2.5) (median (IQR)) SAMIRIX by experienced raters (FCO, CB and SS for ring Average age at onset – 32.9 (24.9–42.4) 30.0 (17.6–42.5) scans, HZ, FCO and AL for macular scans) at a single site at the (years, median (IQR)) Charité—Universitätsmedizin Berlin. To assure comparability Patients with a history – 142 (72.1) 24 (75.0) with previously published data on ORL changes in NMOSD, of ON the macular volume data were further segregated into one of (N (%)) three export protocols: (1) a 5 mm diameter cylinder omitting Median number of ON – 1.00 (0.00–3.00) 2.00 (1.00–4.00) episodes (median, IQR) a 1 mm diameter around the fovea (5 mm study), (2) a 3 mm Disease duration – 7.1±6.7 4.8±7.8 diameter cylinder omitting a 1 mm diameter around the fovea (3 (years, mean±SD) mm study) and (3) a 1 mm mean thickness around the fovea (1 Ethnicity White (57 (761)) White (105 (53.3) White (19 (59.4)) mm study). Results are reported for the 5 mm study on Spectralis (N (%)) Asian (16 (21.3)) Asian (56 (28.4)) Asian (13 (40.6)) Hispanic (1 (1.3)) African American (11 (5.6)) devices; confirmatory results based on the 3 mm and 1 mm study Other (1 (1.3)) Other (25 (12.7)) as well as for Cirrus and Topcon devices are set out in the online Current treatment Rituximab (51 (25.9)) Rituximab (6 (18.8)) supplemental file 1. (N (%)) Azathioprine (42 (21.3)) Azathioprine (6 (18.8)) Mycophenolate Mofetil Prednisone (6 (18.8)) (31 (15.7)) Mycophenolate Methotrexate (4 (2.0)) mofetil (5 (15.6)) Statistical methods Other or missing (69 (35.0)) Other or missing (9 (28.1)) Data were stratified in cohorts by (1) antibody status and (2) ON OCT device (N (%)) Spectralis (75 Spectralis (139 (70.6)) Spectralis (25 (78.1)) history (contralateral eyes of patients with a history of unilateral (100)) Cirrus (38 (19.3)) Cirrus (3 (9.4)) ON are classified not fulfilling the ON history criteria). The data Topcon (20 (10.2)) Topcon (4 (12.5)) were further bifurcated by OCT device (Spectralis, Cirrus or Cirrus: Cirrus HD-­OCT, Carl Zeiss Meditec Inc, Dublin, California, USA; Spectralis: SD-­OCT, Heidelberg Engineering, Heidelberg, Germany;Topcon: Topcon 3D-­OCT, Topcon Corp, Tokyo Japan. Topcon) to mitigate any device-­specific aberrations. For contin- AQP4-­IgG, anti-­aquaporin-­4 antibody; EDSS, Expanded Disability Standard Scale; HCs, healthy controls; uous cohort data (age, average age at onset and disease dura- MOG-­IgG, anti-­myelin-­oligodendrocyte-­glycoprotein antibody; N, number of subjects; ON, optic neuritis. tion) on each of the AQP4-­IgG, MOG-­IgG and HC cohorts, the Student’s t-­test was employed. Cross-­sectional group compar- APOSTEL V.2.0 recommendations.24 Peripapillary retinal nerve isons of the OCT values were conducted using linear mixed-­ fibre layer (pRNFL) thickness was derived using a device-­specific effect models with age and sex as fixed and centre and patient-­ID protocol and centred around the optic nerve head. Segmentation as random effects; where necessary, models were corrected for of all layers in macular volume scans were performed semiauto- age and sex. Marginal and conditional coefficients of determi- matically and processed with an in-­house proprietary software nation for the models were estimated by pseudo-­R2 for mixed-­ (SAMIRIX).25 For the purposes of this study, the macular retinal effect models. Significance was established at p<0.05. Statistical Table 2 Group comparison between HC and patients who were AQP4-­IgG and MOG-­IgG seropositive at baseline (Spectralis devices only) AQP4-­IgG vs HC AQP4-­IgG vs MOG-­IgG MOG-­IgG vs HC HC AQP4-­IgG MOG-­IgG B SE P B SE P B SE P Number of eyes 148 317 55 pRNFL in 99.17±9.76 78.46±24.13 74.33±23.44 −20.22 2.86 <0.001 0.34 4.33 0.937 −29.40 2.75 <0.001 µm (mean±SD) mRNFL in 35.25±3.13 28.09±6.60 27.62±5.43 −6.12 0.69 <0.001 −0.15 1.38 0.913 −6.98 0.66 <0.001 µm (mean±SD) GCIP in 80.62±6.14 65.81±13.03 66.16±11.85 −14.74 1.45 <0.001 −2.18 2.95 0.461 −15.16 1.33 <0.001 µm (mean±SD) INL in 39.64±2.51 39.85±3.57 41.55±4.14 0.34 0.39 0.384 −1.93 0.87 0.028 1.79 0.53 0.001 µm (mean±SD) OPL in 24.58±1.64 25.02±2.03 25.10±2.00 0.28 0.24 0.241 −0.21 0.44 0.634 −0.01 0.29 0.986 µm (mean±SD) ONL in 63.59±5.78 61.63±7.04 64.71±7.87 −0.01 0.83 0.993 −1.77 1.80 0.327 0.69 0.93 0.457 µm (mean±SD) OPNL in 89.23±6.95 86.65±7.21 89.81±8.61 −0.41 0.85 0.634 −1.54 1.85 0.406 −0.14 0.93 0.878 µm (mean±SD) PR in 80.80±2.38 80.35±2.94 81.49±3.59 −0.30 0.33 0.363 −0.07 0.68 0.923 0.20 0.39 0.610 µm (mean±SD) RT in 324.47±13.24 300.76±20.11 306.6±17.99 −20.16 2.37 <0.001 −6.61 4.77 0.169 −18.91 2.49 <0.001 µm (mean±SD) AQP4-­IgG, anti-­aquaporin-­4 antibody; B, estimate; GCIP, ganglion cell and inner plexiform layer; HC, healthy control; INL, inner nuclear layer; MOG-­IgG, anti-­myelin-­oligodendrocyte-­glycoprotein antibody; mRNFL, macular retinal nerve fibre layer; ONL, outer nuclear layer; OPL, outer plexiform layer; OPNL, outer plexiform and nuclear layer; PR, photoreceptor layer; pRNFL, peripapillary retinal nerve fibre layer; RT, total retinal thickness. 190 Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 Neuro-­i nflammation Figure 2 Group comparison of HC and patients who were AQP4-­IgG and MOG-­IgG seropositive at baseline: boxplots of mean OCT values with individual eyes (jitter) in HC (left, green), patients with AQP4-­IgG (middle, yellow) and patients with MOG-­IgG (right, blue). (A) pRNFL; (B) GCIP; (C) INL; (D) OPL; (E) ONL; and (F) PR. AQP4, aquaporin-­4; HC, healthy control; GCIP, ganglion cell and inner plexiform layer; INL, inner nuclear layer; MOG, myelin-­ oligodendrocyte-­glycoprotein; OCT, optical coherence tomography; ONL, outer nuclear layer; OPL, outer plexiform layer; PR, photoreceptive layer; pRNFL, peripapillary retinal nerve fibre layer analyses were conducted using R (V.4.0.0) (RStudio Inc, Boston, AQP4-­IgG seropositive eyes with a history of ON (AQP4-­ON) Massachusetts, USA).27 did not display any thinning of ONL and OPL compared with patients without a history of ON (AQP4-­NON) or HC, despite RESULTS severe neuroaxonal loss measured by pRNFL and GCIP layer Cohort description (table 3, figure 3). Comparing patients who were AQP4-­IgG and In total, 197 patients who were AQP4-­IgG seropositive fulfilled MOG-­IgG seropositive, both groups had a comparable neuro- the inclusion criteria (figure 1, table 1). We also included 75 axonal loss (pRNFL, GCIP)—in the whole group as well as in unmatched HCs and 32 patients who were MOG-­IgG seroposi- respect of ON and non-­ON eyes (table 2, figure 2). AQP4-­ON tiveas control groups. (B=−1.54, SE=0.69 µm, p=0.027) as well as MOG-­ ON Neuroaxonal damage measured by pRNFL, mRFNL and (B=−2.51, SE=0.87 µm, p=0.004) showed an OPL thinning in GCIP was comparable in patients who were AQP4-­IgG sero- the fovea (1 mm diameter) compared with HC, but no difference positive (pRNFL: 78.46±24.13 µm, mRNFL: 28.09±6.60 was observed between AQP4-­ON and MOG-­ON (p=0.100). µm, GCIP: 65.81±13.03 µm) and MOG-­ IgG seropositive Also, no significant correlation between ethnicity and current (pRNFL: 74.33±23.44 µm, mRNFL: 27.62±5.43 µm, GCIP: therapies on outer retinal thickness was found (data not shown). 66.16±11.85 µm) making MOGAD a highly relevant compara- tive disease control group for our investigation of ORLs (table 2). DISCUSSION Limited outer retinal changes in AQP4-IgG seropositive Our study suggests that neither macular OPL nor ONL loss NMOSD occurs in AQP4-­IgG seropositive NMOSD, regardless of ON No significant thinning of macular OPL and ONL in patients phenotype, as compared with HC and patients who were MOG-­ who were AQP4-­IgG seropositive (irrespective of ON status) IgG seropositive. The MOG-­ IgG cohort presented a unique were observed compared with HC or patients who were MOG-­ opportunity to contrast our AQP4-­IgG seropositive cohort with IgG seropositive using the 5 mm diameter macular data (table 2, a highly relevant comparator group, which most likely has no figure 2). No significant changes were observed when the OPL astrocytopathy-­component.28 and ONL values were analysed as the combined OPNL. Previous Our results differ from those published by You et al in 20198 and studies described ORL thinning only in the foveal and parafoveal Filippatou et al in 2020.11 In both studies, thinning was observed area as a sign of AQP4-­IgG-­induced Müller cell damage.8 11 We in the ONL and the inner segment of the photoreceptor layers. In therefore repeated our analyses in both 3 mm and the 1 mm the case of You et al, who utilised Spectralis SD-­OCT devices for diameter volumes around the fovea, but these narrower volumes the image acquisition, foveal thinning was observed along with a showed again no relevant OPL or ONL thinning in patients who reduction in b-­wave amplitudes in full-­field electroretinography were AQP4-­IgG seropositive compared with HC or patients who (ERG) suggestive of Müller cell dysfunction.8 Filippatou et al, were MOG-­ IgG seropositive (see online supplemental data). who employed Cirrus-­SD-­OCT for the image acquisition, also Additionally, while these previous studies reported changes in described thinning of the fovea in the 5 mm diameter macular the inner segment layer of the photoreceptors, this was not seen area around the fovea.11 Both studies suggested the ORL changes in our study.8 11 to be caused by a primary retinal astrocytopathy with AQP4-­IgG After a previous description11 of ORL changes in patients who associated glial dysfunction in Müller cells.29 These pathological were AQP4-­IgG seropositive with a history of ON, we also exam- responses could account for the associated thinning observed ined ORL differences separately in eyes with a history of ON. in the ONL in these studies. However, other exogenous factors Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 191 Neuro-­i nflammation cannot be ruled out as contributory, such as cohort composition 0.275 0.344 0.899 0.446 0.450 0.348 AQP4, aquaporin-­4 ; B, estimate; GCIP, ganglion cell and inner plexiform layer; HC, healthy control; MOG, myelin-­oligodendrocyte-­glycoprotein; NON, non-­optic neuritis; ON, optic neuritis; ONL, outer nuclear layer; OPL, outer plexiform layer; OPNL, outer plexiform and nuclear layer; PR, photoreceptor and study methodologies. AQP4-­NON vs MOG-­NON P On a cohort level, our population is larger (197 patients who were AQP4-­IgG seropositive vs 22 and 51 by You et al and Filip- 5.44 3.36 0.96 3.58 3.51 1.44 patou et al, respectively)8 11 and more diverse than prior studies, SE which minimises potential type I errors. While You et al did 6.03 3.22 2.76 2.67 1.36 −0.12 not specify the ethnic composition of their cohort, the cohort in Filippatou et al had a relatively even distribution between B Caucasian Americans (43%) and African Americans (53%) 0.667 0.215 0.183 0.509 0.259 0.454 with a minor subset of Asian Americans (4%)—describing a P AQP4-­ON vs MOG-­ON pronounced ONL thinning in African Americans. African Amer- ican patients with multiple sclerosis (MS) are also known to suffer 6.12 3.20 2.12 0.51 2.21 0.77 SE from faster and often more aggressive disease course in general, which could also be true for other neuroinflammatory diseases 2.63 0.34 −4.00 −2.84 −2.51 −0.58 like NMOSD.30 31 Our AQP4-­IgG seropositive cohort included B an ethnically diverse dataset acquired worldwide with a lower African American patient composition (5.6%), which might have 0.735 0.804 0.560 0.794 0.071 0.07 contributed to the less profound foveal ONL changes.11 32 P Recently, it has been hypothesised that the neuroplastic char- AQP-­NON vs HC 5.15 1.48 0.38 1.40 1.42 0.59 acteristics of the INL may act as a barrier to retrograde (but not SE anterograde) trans-­ synaptic axonal degeneration—rectified to the ORLs—in patients with MS following ON.33 This limited 0.09 0.37 −9.29 −0.50 −0.82 −1.08 neuroplastic ability is hypothesised to rest with the bipolar, B amacrine and horizontal cells, which feed into the synaptic tree 0.184 0.847 0.775 0.595 <0.001 <0.001 at the level of the INL, and raises questions as to whether such protective mechanisms may also play a limited part in NMOSD P and whether it remains so as we age.33 The average age of partic- 3.57 1.49 0.26 0.92 0.95 0.36 ipants in the two other studies were relatively older (mean age AQP4-­ON vs HC SE OCT results in patients who were AQP4-­IgG seropositive stratified by history of on (Spectralis devices only) for both being 47 years), whereas for our AQP4-­IgG cohort it was 42 years. Previously reported studies concerning cohorts of 0.34 0.27 −0.18 −0.19 −29.56 −19.60 similar demographic distribution to ours reported no significant B correlation between age and retinal thickness.34 35 However, age-­ related changes in the retina cannot be ruled out and ORLs may 0.498 0.901 0.879 0.187 <0.001 <0.001 be more susceptible to change with increasing age and/or disease P AQP4-­ON vs AQP4-­NON duration. It is well-­known that the plasticity of the CNS mark- edly reduces over time, and as a corollary, the regenerative prop- 3.93 2.06 0.38 1.49 1.53 0.57 SE erties of the INL may also be affected thereby diminishing its protective effects in reducing retrograde (trans-­synaptic) axonal 0.26 0.00 0.75 −0.19 −25.18 −14.74 degeneration.36 The retina is also a vascularised organ, particu- larly at the interface between inner and outer retina, where the B deep vascular plexus intercepts the boundary between the INL and OPL.37 Should the blood–retina barrier be compromised 95.33±7.32 75.88±6.01 24.45±1.55 59.76±4.58 91.37±8.69 79.62±3.58 MOG-­NON in the boundary between the INL and OPL, it is conceivable 12 that the protective abilities of the INL may be circumvented and thereby mediating glial dysfunction in the Müller cells. This may have been what was observed in the OPL from the 1 68.03±22.95 63.45±11.96 25.28±2.08 66.09±8.08 84.21±5.68 82.01±3.45 MOG-­ON mm AQP4-­ON and MOG-­ON cohort given the relative loca- tion of the OPL to the INL. To that end, while disease duration 43 did not reveal to any correlates with OPL (p=0.805) or ONL (p=0.835) values, we cannot exclude time-­dependent effects in a 96.09±12.99 AQP4-­NON 77.11±7.56 24.71±1.79 63.14±6.62 87.85±6.78 79.80±2.94 cross-­sectional analysis. We believe that this area warrants more research to quantify if (1) age is a factor, (2) ON damages the 85 barrier function and (3) the INL does indeed play a role as a dam layer; pRNFL, peripapillary retinal nerve fibre layer. to retrograde axonal degeneration in NMOSD. A strength of our study rests on its cohort size and compo- 72.84±24.47 62.94±12.73 25.06±2.01 62.53±7.45 87.58±7.67 80.89±2.93 AQP4-­ON sition, which mirrors that of a global population. This result derives from a consortium of expert NMOSD researchers 232 enabling the enrolment of participants through a multicentre pRNFL in µm (mean±SD) strategy. This approach was designed to overcome many of the earlier NMOSD study limitations, for example small and homo- Number of eyes geneous sample populations. Additionally, the use of differing µm (mean±SD) µm (mean±SD) µm (mean±SD) µm (mean±SD) µm (mean±SD) Table 3 OCT devices compounds complexities in OCT comparisons and OPNL in GCIP in ONL in OPL in a high degree of caution is needed in order to rely on differing PR in platforms interchangeably.38 Thus, our study focuses on use of 192 Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 Neuro-­i nflammation Figure 3 OCT results stratified by ON status (tested with Spectralis devices): boxplots of mean OCT values with individual eyes (jitter) in HC (left, green), AQP4-­IgG cohort (middle) and MOG-­IgG cohort (right). Seropositive patients with a history of ON are highlighted with light yellow and seropositive patients without a history of ON are highlighted in orange. (A) pRNFL; (B) GCIP; (C) INL; (D) OPL; (E) ONL; and (F) PR. AQP4, aquaporin-­4; GCIP, ganglion cell and inner plexiform layer; INL, inner nuclear layer; MOG, myelin-­oligodendrocyte-­glycoprotein; OCT, optical coherence tomography; ONL, outer nuclear layer; OPL, outer plexiform layer; PR, photoreceptive layer; pRNFL, peripapillary retinal nerve fibre layer. three widely available OCT devices, and obtained confirmatory Author affiliations 1 results with each of them; of these, two were also employed Experimental and Clinical Research Center, Max Delbrück Center for Molecular Medicine and Charité – Universitätsmedizin Berlin, corporate member of Freie respectively in the studies by You et al8 and Filippatou et al.11 Universität Berlin and Humboldt-­Universität zu Berlin, Berlin, Germany Limitations of the current study should also be considered. 2 NeuroCure Clinical Research Center, Charité – Universitätsmedizin Berlin, corporate First, the HCs and patients with MOGAD were not matched, member of Freie Universität Berlin and Humboldt-­Universität zu Berlin, Berlin, which makes it difficult to rule out age-­ related and gender-­ Germany 3 related affects. Notably, retinal thickness decreases with age and CIEM MS Research Center, University of Minas Gerais State, Medical School, Belo Horizonte, Brazil males generally exhibit higher GCIP and RT.25 Also, no ERG or 4 Kashani MS Center, Isfahan University of Medical Sciences, Isfahan, Iran (the Islamic functional visual pathway assessments were conducted, which Republic of) 5 could have potentially shown more subtle functional impairment School of Advanced Technologies in Medicine, Medical Image and Signal Processing of ORLs without associated tissue loss. Outer retinal studies are Research Center, Isfahan University of Medical Sciences, Isfahan, Iran (the Islamic Republic of) additionally complicated by Henle Fibre morphologies as OCT 6 Isfahan Eye Research Center, Department of Ophthalmology, Isfahan University of beam placement plays a major role in how this layer is depicted; Medical Sciences, Isfahan, Iran (the Islamic Republic of) the high level of irregularity and variability in these morphol- 7 Center for Advanced Neurological Research, Nitte University, Mangalore, Karnataka, ogies add a level of subjectiveness in the quantification and India 8 correction of outer layer segmentation and analyses.39 Finally, Department of Neurology, National Cancer Center Korea, Goyang-­si, Korea (the Republic of) Cirrus and Topcon measurements could not be utilised as confir- 9 Department of Opthalmology, Research Institute and Hospital of National Cancer matory cohorts as there lacked sufficient HCs examined with Center, Goyang, Korea (the Republic of) these devices. Nonetheless, the current findings provide insights 10 Experimental Neurophysiology Unit, Institute of Experimental Neurology (INSPE) into relationships between retinal layer changes and axonal Scientific Institute, Hospital San Raffaele and University Vita-­Salute San Raffaele, damage that have not previously been recognised; as no ORL Milano, Italy 11 Division of Neurology, Department of Medicine, Siriraj Hospital and Bumrungrad changes can be observed on account of a primary astrocytopathy International Hospital, Bangkok, Thailand in NMOSD, it potentially alleviates the burden of monitoring 12 Swedish Neuroscience Institute Neuro-­Ophthalmology, Seattle, Washington, USA 13 the ORLs when tracking disease progression and reinforces Department of Neurology, Neurology Service, University Hospital of Strasbourg, the need to focus primarily on the inner layers, particularly the Strasbourg, France 14 Department of Neurology, Oxford University Hospitals NHS Trust, Oxford, UK RNFL and the GCIP layer. 15 Department of Ophthalmology, Oxford University Hospitals NHS Trust, Oxford, UK 16 Neuro-­Opthalmology Division, Department of Opthalmology, Rabin Medical Center, Petah Tikva, Israel 17 CONCLUSION Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel 18 Our results show no evidence of macular ORL changes as a major Department of Neurology Slagelse, Institutes of Regional Health Research andMolecular Medicine, University of Southern Denmark, Odense, Syddanmark, component of retinal damage in patients who were seropositive Denmark AQP4-­IgG NMOSD and patients with MOGAD. Further studies 19 Institute of Regional Health Research, University of Southern Denmark, Odense, will be necessary to clarify (1) if OPL and ONL are damaged in Denmark 20 late disease stages due to retrograde trans-­synaptic axonal degen- Hospital Clinic of Barcelona-­Institut d’Investigacions, Biomèdiques August Pi eration across the damaged INL barrier and (2) if outer retinal Sunyer, University of Barcelona, Barcelona, Spain 21 Institute of Clinical Neuroimmunology, LMU Hospital, Ludwig-­Maximilians-­ dysfunction without a measurable structural correlate occurs. Universitat Munchen, Munich, Germany Longitudinal studies could help quantify changes in the ORLs 22 Department of Neurology, University of Michigan Medical School, Ann Arbor, alongside disease progression. Michigan, USA Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 193 Neuro-­i nflammation 23 NYU Multiple Sclerosis Comprehensive Care Center, Department of Neurology, and speaker honoraria from Merck Serono, Pacific Healthcare (Thailand), Menarini NYU, New York, New York, USA (Thailand), Biogen Idec, UCB (Thailand), and Novartis. AA reports personal fees 24 Neurology, Multiple Sclerosis, Myelin Disorders and Neuroinflammation, Hospital from received honoraria for giving educational presentations on multiple sclerosis for Neurology Pierre Wertheimer, Lyon, France and neuroimmunology at several national congresses or symposia from Teva 25 Centre d’Esclerosi Múltiple de Catalunya (Cemcat). Department of Neurology/ Turkey, Merck-­Serono, Biogen Idec-­Gen Pharma of Turkey, Roche, Novartis, Bayer, Neuroimmunology, Universitat Autònoma de Barcelona, Barcelona, Spain Sanofi-­Genzyme. She has received travel and registration coverage for attending 26 Department of Neurology, Koc University Research Center for Translational several national and international congresses or symposia from Merck-­Serono, Medicine (KUTTAM), Koc University School of Medicine, Istanbul, Turkey Biogen Idec-­Gen Pharma of Turkey, Roche, Sanofi-­Genzyme and Bayer. AJ has 27 Cerrahpaşa Faculty of Medicine, Department of Neurology, Istanbul University-­ received compensation for advisory board, consulting, meeting attendance and Cerrahpaşa, Istanbul, Turkey speaking from Biogen, Terumo-­BCT, Genentech, Shire and Chugai Pharmaceuticals. 28 Department of Ophthalmology, Cerrahpasa Medical Faculty, Istanbul Universitesi, SH has received funding from the NMO Spectrum-­UK charity and was previously Fatih, Turkey funded by an MGA/Watney/NIHR Oxford Biomedical research grant. RM serves on 29 Department of Neurology, Medical Faculty, Heinrich-­Heine-­Universitat Dusseldorf, scientific advisory board for MedImmune and has received funding for travel and Dusseldorf, Nordrhein-­Westfalen, Germany honoraria from Biogen, Merck Serono, Novartis, Sanofi- Genzyme, Roche and Teva. 30 Department of Neurology, Center for Neurology and Neuropsychiatry, LVR-­ EN has nothing to disclose. ACC received funding from the Instituto de Salud Carlos Klinikum, Heinrich-­Heine-­Universitat Dusseldorf, Dusseldorf, Germany III (Spain) JR19/00007 unrelated to this manuscript. DB has received speaking/ 31 Department of Ophthalmology and Visual Sciences, Escola Paulista de Medicina, consulting honoraria from Bayer Health Care, Biogen Idec, Merck, Sanofi-­Genzyme, Universidade Federal de São Paulo, São Paulo, Brazil TEVA and Roche and had travel expenses to scientific meetings sponsored by Bayer 32 Department of Neurology and Neurosurgery, Escola Paulista de Medicina, Health Care, Merck Serono, TEVA and Roche. JH reports grants for OCT research Universidade Federal de São Paulo, Sao Paulo, Brazil from the Friedrich-­Baur-­Stiftung and Merck, personal fees and non-­financial support 33 The Walton Centre NHS Foundation Trust, Liverpool, UK from Celgene, Merck, Alexion, Novartis, Roche, Santhera, Biogen, Heidelberg 34 Department of Neurology, Hospital Clinico de Maracaibo, Maracaibo, Venezuela, Engineering, Sanofi Genzyme and non-­financial support of the Guthy-­Jackson Bolivarian Republic of Charitable Foundation, all outside the submitted work. JH is partially funded by the 35 Moorfield’s Eye Hospital, The National Hospital for Neurology and Neurosurgery, German Federal Ministry of Education and Research (DIFUTURE), Grant Numbers Queen Square Institute of Neurology, University College London, London, UK 01ZZ1603[A-­D] and 01ZZ1804[A-­H]. LL received honoraria for consulting services 36 Department of Neurology, University of California San Francisco, San Francisco, from Merck, Roche, Biogen and for speaking activities from Teva; research support California, USA from Merck, Biogen, Novartis; travel support from Merck, Roche, Biogen, Almirall. 37 MP has nothing to disclose. OA has received honoraria for speaking/consultation Department of Medicine, Harbor-­University of California at Los Angeles (UCLA) Medical Center, and Lundquist Institute for Biomedical Innovation, Torrance, and travel grants from Bayer Healthcare, Biogen Idec, Chugai, Novartis, Medimmune, California, USA Merck Serono, and Teva and research grants from Bayer Healthcare, Biogen Idec, 38 Novartis, and Teva. MR received speaker honoraria from Novartis, Bayer, Roche, Department of Medicine, David Geffen School of Medicine, UCLA, Los Angeles, California, USA Alexion and Ipsen and travel reimbursement from Bayer, Biogen, Merz, Genzyme, 39 Departments of Ophthalmology and Visual Sciences, Kellogg Eye Center, Ann Arbor, Teva, Roche and Merck, none related to this study. PA reports grants, personal fees Michigan, USA and non-­financial support from Allergan, Biogen, Ipsen, Merz Pharmaceuticals, 40 Department of Metabolism, Endocrine and Diabetes, Department of Internal Novartis, and Roche, personal fees and non-­financial support from Bayer Healthcare, Medicine, University of Michigan Medical School, Ann Arbor, Michigan, USA and Merck, and non-­financial support from Sanofi-­Aventis/Genzyme. HJK reports 41 Department of Pediatrics, University of Utah Health, Salt Lake City, Utah, USA speaking and/or consulting: Bayer Schering Pharma, Biogen, Celltrion, Eisai, HanAll 42 Department of Neurology, Charité – Universitätsmedizin Berlin, corporate member BioPharma, MedImmune, Merck Serono, Novartis, Sanofi Genzyme, Teva-­Handok, of Freie Universität Berlin and Humboldt-­Universität zu Berlin, Berlin, Germany and UCB; research support: Ministry of Science & ICT, Sanofi Genzyme, Teva-­Handok, 43 Department of Neurology, University of California Irvine, Irvine, California, USA and UCB; steering committee member: MedImmune; co-­editor/associated editor: MS Journal-­Experimental, Translational and Clinical; and Journal of Clinical Neurology. J-­WH has received a grant from the National Research Foundation of Korea. YM-­D Twitter Elena H Martinez-­Lapiscina @elenahlapiscina@ and Joachim Havla @ has served as a consultant and/or received grant support from: Acorda, Bayer NeuroVisionLab Pharmaceutical, Biogen Idec, Celgene, EMD Serono, Genzyme, Novartis, Questor, Contributors The study was conceived and designed by AL, HZ, FCO, AB and FP. Chugai, and Teva Neuroscience and is currently supported by grants from NIH NIAID AL, HZ, FCO, CB, SM and SvS collected and analysed data for the study. HZ, FCO, Autoimmune Center of Excellence: UM1-­AI110557; NIH NINDS R01-­NS080821. AB and FP contributed to the scientific design and conduct of the study. AL and HSK has nothing to disclose. IK served on scientific advisory board for Biogen Idec FCO performed biostatistical analyses to the study, were responsible for literature and Genentech and received research support from Guthy-­Jackson Charitable research and wrote the manuscript. AP, AJG, MRY, LC and TJS contributed to the Foundation, National Multiple Sclerosis Society, Biogen-­Idec,Serono, Genzyme and conception and design of the study. All other authors contributed to the acquisition Novartis. ZR has nothing to disclose. AR has nothing to disclose. MRY is founder and and analysis of data. All authors approved the final version of the manuscript. a shareholder of NovaDigm Therapeutics, Inc; he receives funding from the United Funding The authors acknowledge support from the Guthy Jackson Charitable States National Institutes of Health and United States Department of Defense; Foundation (GJCF) and the German Research Foundation (DFG) pertaining to the he holds US and international patents on immunotherapeutic and anti-­infective CROCTINO project. technologies, is a member of the Genentech-­Roche Scientific Advisory Committee and adviser to The Guthy-­Jackson Charitable Foundation. TJS was issued US patents Competing interests HZ reports grants from Novartis and speaking honoraria covering the therapeutic targeting of IGF-­I receptor in autoimmune diseases. He is from Bayer Healthcare, unrelated to this study. EHM-­L received funding from the a paid consultant for Horizon Thera and Immunovant and is a scientific advisor to Instituto de Salud Carlos III (Spain) and Fondo Europeo de Desarrollo Regional the Guthy-­Jackson Charitable Foundation. He receives research funding from the (FEDER-­JR16/00006), Grant for MS Innovation, Fundació Privada Cellex and National Institutes of Health. AP is supported by the National Institute for Health Marató TV3 Charitable Foundation and is a researcher in the OCTIMS study, an Research (NIHR) Biomedical Research Centre based at Moorfields Eye Hospital observational study (that involves no specific drugs) to validate SD-­OCT as a National Health Service (NHS) Foundation Trust and University College London biomarker for MS, sponsored by Novartis and has received honoraria and travel Institute of Ophthalmology. AB is cofounder and shareholder of Motognosis and support for international and national meetings over the last 3 years from from Nocturne. He is named as inventor on several patent applications regarding MS Biogen, Novartis, Roche, Genzyme. She is a member of the working committee of serum biomarkers, OCT image analysis and perceptive visual computing. FP reports International Multiple Sclerosis Visual System (IMSVISUAL) Consortium. MAL-­P research grants and speaker honoraria from Bayer, Teva, Genzyme, Merck, Novartis, has received funding for travel and speaker honoraria from Novartis, Sanofi- MedImmune and is member of the steering committee of the OCTIMS study Genzyme and Roche. MAF has nothing to disclose. Jacqueline Palace has received (Novartis), all unrelated to this work. FCO was employee of Nocturne GmbH and support for scientific meetings and honorariums for advisory work From Merck receives research support by the American Academy of Neurology and National Serono, Novartis, Chugai, Alexion, Roche, Medimmune, Argenx, UCB, Mitsubishi, Multiple Sclerosis Society (US), unrelated to this work as well as funding by the Amplo, Janssen. Grants from Alexion, Amplo biotechnology. Shares in AstraZenica. German Association of Neurology (Deutsche Gesellschaft für Neurologie) in context Acknowledges Partial funding by Highly specialised services NHS England. MIL of this project.CC has received a speaking honorarium from Bayer and research reported being involved in aquaporin 4 testing, receiving salary from the National funding from Novartis unrelated to this publication. All other authors have nothing Health Service National Highly Specialised Commissioning Group for Neuromyelitis to disclose. Optica, UK, being supported by the National Institute for Health Research Oxford Patient consent for publication Consent obtained directly from patient(s). Biomedical Research Centre, UK, and receiving speaking honoraria and travel grants from Biogen Idec, and travel grant from Novartis. SMS has nothing to Ethics approval Written informed consent was obtained from all patients prior disclose. AR-­F is sponsored by Abide Therapeutic outside of the submitted work to the commencement of the study and institutional review board approvals for and reports no potential conflicts of interest. SSiritho received funding for travel retrospective data use were obtained or waived from each centre in accordance with 194 Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 Neuro-­i nflammation the Declaration of Helsinki (1964) in its currently applicable version at their own 15 Oertel FC, Specovius S, Zimmermann HG, et al. Retinal optical coherence tomography discretion and in accordance with relevant local laws. The study also conformed to all in neuromyelitis optica. Neurol Neuroimmunol Neuroinflamm 2021;8. doi:10.1212/ relevant best practice guidelines and ethical standards of each centre. NXI.0000000000001068. [Epub ahead of print: 15 Sep 2021]. 16 Bruijstens AL, Wong YYM, van Pelt DE, et al. Hla association in MOG-­IgG- and AQP4-­ Provenance and peer review Not commissioned; externally peer reviewed. IgG-­related disorders of the CNS in the Dutch population. Neurol Neuroimmunol Data availability statement Data are available upon reasonable request. Neuroinflamm 2020;7. doi:10.1212/NXI.0000000000000702. [Epub ahead of print: Supplemental material This content has been supplied by the author(s). It 20 Mar 2020]. has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have 17 Kim H, Lee E-­J, Kim S, et al. Serum biomarkers in myelin oligodendrocyte been peer-­reviewed. Any opinions or recommendations discussed are solely those glycoprotein antibody-­associated disease. Neurol Neuroimmunol Neuroinflamm of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and 2020;7. doi:10.1212/NXI.0000000000000708. [Epub ahead of print: 17 Mar responsibility arising from any reliance placed on the content. Where the content 2020]. includes any translated material, BMJ does not warrant the accuracy and reliability 18 Narayan R, Simpson A, Fritsche K, et al. Mog antibody disease: a review of MOG of the translations (including but not limited to local regulations, clinical guidelines, antibody seropositive neuromyelitis optica spectrum disorder. Mult Scler Relat Disord terminology, drug names and drug dosages), and is not responsible for any error 2018;25:66–72. and/or omissions arising from translation and adaptation or otherwise. 19 Havla J, Kümpfel T, Schinner R, et al. Myelin-­oligodendrocyte-­glycoprotein (MOG) autoantibodies as potential markers of severe optic neuritis and subclinical retinal Open access This is an open access article distributed in accordance with the axonal degeneration. J Neurol 2017;264:139–51. Creative Commons Attribution Non Commercial (CC BY-­NC 4.0) license, which 20 Reindl M, Schanda K, Woodhall M, et al. International multicenter examination of permits others to distribute, remix, adapt, build upon this work non-­commercially, MOG antibody assays. Neurol Neuroimmunol Neuroinflamm 2020;7:e674. and license their derivative works on different terms, provided the original work is 21 Waters P, Reindl M, Saiz A, et al. Multicentre comparison of a diagnostic assay: properly cited, appropriate credit is given, any changes made indicated, and the use aquaporin-­4 antibodies in neuromyelitis optica. J Neurol Neurosurg Psychiatry is non-­commercial. See: http://creativecommons.org/licenses/by-nc/4.0/. 2016;87:1005–15. 22 Tewarie P, Balk L, Costello F, et al. The OSCAR-­IB consensus criteria for retinal OCT ORCID iDs quality assessment. PLoS One 2012;7:e34823. Angelo Lu http://orcid.org/0000-0002-7897-6498 23 Schippling S, Balk LJ, Costello F, et al. Quality control for retinal OCT in multiple Claudia Chien http://orcid.org/0000-0001-8280-9513 sclerosis: validation of the OSCAR-­IB criteria. Mult Scler 2015;21:163–70. Ho Jin Kim http://orcid.org/0000-0002-8672-8419 24 Aytulun A, Cruz-­Herranz A, Aktas O, et al. Apostel 2.0 recommendations for reporting Adriana Roca-­Fernández http://orcid.org/0000-0002-8720-9397 quantitative optical coherence tomography studies. Neurology 2021;97:68–79. Hadas Stiebel-­Kalish http://orcid.org/0000-0001-7715-6706 25 Motamedi S, Gawlik K, Ayadi N, et al. Normative data and minimally detectable Elena H Martinez-­Lapiscina http://orcid.org/0000-0003-4272-0826 change for inner retinal layer thicknesses using a semi-­automated OCT image Joachim Havla http://orcid.org/0000-0002-4386-1340 segmentation pipeline. Front Neurol 2019;10:1117. Marius Ringelstein http://orcid.org/0000-0003-3618-8407 26 Staurenghi G, Sadda S, Chakravarthy U, et al. Proposed lexicon for anatomic Philipp Albrecht http://orcid.org/0000-0001-7987-658X landmarks in normal posterior segment spectral-­domain optical coherence Denis Bernardi Bichuetti http://orcid.org/0000-0002-4011-3734 tomography: the IN•OCT consensus. Ophthalmology 2014;121:1572–8. Axel Petzold http://orcid.org/0000-0002-0344-9749 27 R Core Team. R: a language and environment for statistical computing. Vienna, Frederike Cosima Oertel http://orcid.org/0000-0003-4906-5983 Austria: R Foundation for Statistical Computing, 2017. https://www.R-project.org/ 28 Höftberger R, Guo Y, Flanagan EP, et al. The pathology of central nervous system REFERENCES inflammatory demyelinating disease accompanying myelin oligodendrocyte 1 Jarius S, Paul F, Weinshenker BG, et al. Neuromyelitis optica. Nat Rev Dis Primers glycoprotein autoantibody. Acta Neuropathol 2020;139:875–92. 2020;6:85. 29 Goodyear MJ, Crewther SG, Junghans BM. A role for aquaporin-­4 in fluid regulation in 2 Wingerchuk DM, Lennon VA, Lucchinetti CF, et al. The spectrum of neuromyelitis the inner retina. Vis Neurosci 2009;26:159–65. optica. Lancet Neurol 2007;6:805–15. 30 Caldito NG, Saidha S, Sotirchos ES, et al. Brain and retinal atrophy in African-­ 3 Oertel FC, Kuchling J, Zimmermann H, et al. Microstructural visual system changes Americans versus Caucasian-­Americans with multiple sclerosis: a longitudinal study. in AQP4-­antibody-­seropositive NMOSD. Neurol Neuroimmunol Neuroinflamm Brain 2018;141:3115–29. 2017;4:e334. 31 Mealy MA, Kessler RA, Rimler Z, et al. Mortality in neuromyelitis optica is strongly 4 Uzawa A, Mori M, Kuwabara S. Neuromyelitis optica: concept, immunology and associated with African ancestry. Neurol Neuroimmunol Neuroinflamm 2018;5:e468. treatment. J Clin Neurosci 2014;21:12–21. 32 Liu J, Mori M, Zimmermann H, et al. Anti-­Mog antibody–associated disorders: 5 Papadopoulos MC, Verkman AS. Aquaporin 4 and neuromyelitis optica. Lancet Neurol differences in clinical profiles and prognosis in Japan and Germany. J Neurol 2012;11:535–44. Neurosurg Psychiatry 2021;92:377–83. 6 Nagelhus EA, Ottersen OP. Physiological roles of aquaporin-­4 in brain. Physiol Rev 33 Panneman EL, Coric D, Tran LMD, et al. Progression of anterograde trans-­synaptic 2013;93:1543–62. degeneration in the human retina is modulated by axonal convergence and 7 Bringmann A, Pannicke T, Grosche J, et al. Müller cells in the healthy and diseased divergence. Neuroophthalmology 2019;43:382–90. retina. Prog Retin Eye Res 2006;25:397–424. 34 Tian G, Li Z, Zhao G, et al. Evaluation of retinal nerve fiber layer and ganglion 8 You Y, Zhu L, Zhang T, et al. Evidence of Müller glial dysfunction in patients with cell complex in patients with optic neuritis or neuromyelitis optica spectrum aquaporin-­4 immunoglobulin G-­Positive neuromyelitis optica spectrum disorder. disorders using optical coherence tomography in a Chinese cohort. J Ophthalmol Ophthalmology 2019;126:801–10. 2015;2015:1–6. 9 Jeong IH, Kim HJ, Kim N-­H, et al. Subclinical primary retinal pathology in neuromyelitis 35 Martinez-­Lapiscina EH, Sepulveda M, Torres-­Torres R, et al. Usefulness of optical optica spectrum disorder. J Neurol 2016;263:1343–8. coherence tomography to distinguish optic neuritis associated with AQP4 or MOG in 10 Oertel FC, Outteryck O, Knier B, et al. Optical coherence tomography in myelin-­ neuromyelitis optica spectrum disorders. Ther Adv Neurol Disord 2016;9:436–40. oligodendrocyte-­glycoprotein antibody-­seropositive patients: a longitudinal study. J 36 Balk LJ, Coric D, Knier B, et al. Retinal inner nuclear layer volume reflects inflammatory Neuroinflammation 2019;16:154. disease activity in multiple sclerosis; a longitudinal OCT study. Mult Scler J Exp Transl 11 Filippatou AG, Vasileiou ES, He Y, et al. Evidence of subclinical quantitative retinal Clin 2019;5:205521731987158. layer abnormalities in AQP4-­IgG seropositive NMOSD. Mult Scler 2021;27:1738–48. 37 Campbell JP, Zhang M, Hwang TS, et al. Detailed vascular anatomy of the human 12 Oertel FC, Havla J, Roca-­Fernández A, et al. Retinal ganglion cell loss in neuromyelitis retina by Projection-­Resolved optical coherence tomography angiography. Sci Rep optica: a longitudinal study. J Neurol Neurosurg Psychiatry 2018;89:1259–65. 2017;7:42201. 13 Motamedi S, Oertel FC, Yadav SK, et al. Altered fovea in AQP4-­IgG-­seropositive 38 Bhargava P, Lang A, Al-­Louzi O, et al. Applying an open-­source segmentation neuromyelitis optica spectrum disorders. Neurol Neuroimmunol Neuroinflamm algorithm to different OCT devices in multiple sclerosis patients and healthy controls: 2020;7. doi:10.1212/NXI.0000000000000805. [Epub ahead of print: 23 Jun 2020]. implications for clinical trials. Mult Scler Int 2015;2015:1–10. 14 Specovius S, Zimmermann HG, Oertel FC, et al. Cohort profile: a collaborative 39 Gonzalez Caldito N, Antony B, He Y, et al. Analysis of agreement of Retinal-­Layer multicentre study of retinal optical coherence tomography in 539 patients with thickness measures derived from the segmentation of horizontal and vertical neuromyelitis optica spectrum disorders (CROCTINO). BMJ Open 2020;10:e035397. spectralis OCT macular scans. Curr Eye Res 2018;43:415–23. Lu A, et al. J Neurol Neurosurg Psychiatry 2022;93:188–195. doi:10.1136/jnnp-2021-327412 195