Neuromyelitis optica spectrum disorder (NMOSD) is an autoimmune inflammatory disorder that affects the brain and the spinal cord, most frequently the optic nerve (optic neuritis or ON), and/or the spinal cord (myelitis). Patients with NMOSD experience repeated attacks separated by periods of remission (Lalan et al., 2012). The clinical and radiological semiology of these attacks is often similar (although frequently more severe) to other diseases such as multiple sclerosis (MS), acute disseminated encephalomyelitis (ADEM), idiopathic acute transverse myelitis, idiopathic optic neuritis, inflammatory diseases associated with myelin oligodendrocyte glycoprotein, sarcoidosis, and other systemic autoimmune diseases (Kim et al., 2017). The identification of aquaporin-4 specific autoantibodies (AQP4- IgG) has acquired a crucial role in diagnosing NMOSD (Kim et al., 2017). However, even with the most sensitive assays (i.e. cell-based assay), 10–40% of patients diagnosed with NMOSD are seronegative for serum aquaporin-4 immunoglobulin-G antibody (Mariotto et al., 2017). Diagnosis becomes difficult in the absence of these autoantibodies in the patients’ serum, with much more stringent criteria needed for diagnosis as seronegative NMOSD according to the 2015 Wingerchuk consensus diagnostic criteria for NMOSD (Bertsch-Gout et al., 2018). This is further complicated by the lack of availability of the appropriate AQP4-Ab assays in many centers (Bertsch-Gout et al., 2018). MS is the main entity in the differential diagnosis (Dutra et al., 2018) and the correct diagnosis of NMOSD is important because of a different pathophysiology and response to treatments as compared to MS, with possible worsening of NMOSD if inappropriately treated with some MS medications. Therefore, new NMOSD-specific biomarkers are needed to improve the differentiation of NMOSD from MS and other diseases, to enable the physician to initiate timely treatment.

Spectral-domain optical coherence tomography (SD-OCT) has been extensively used in quantifying neuro-axonal damage through the noninvasive in-vivo thickness measurement of the optic nerve and the macula (Mateo et al., 2017; Outteryck and Vermersch, 2016). Indeed, MS and NMOSD patients with optic neuritis have a significantly reduced peri-papillary retinal nerve fiber layer (pRNFL) and macular thicknesses (Outteryck and Vermersch, 2016). Furthermore, these measures were reduced to a greater extent in NMOSD (average pRNFL loss of 35 μ more than in MS) (Outteryck and Vermersch, 2016; Schneider et al., 2013; Vabanesi et al., 2019). Similarly, the ganglion cell inner plexiform layer thickness (GCIPL) was more severely affected in NMOSD than in MS patients (Schneider et al., 2013).

In clinical practice we often encounter subjects with repeatedly negative AQP4- IgG titers who do not satisfy the 2015 Wingerchuk diagnostic consensus criteria for NMOSD (Wingerchuk et al., 2015), but clinically behave like NMOSD. These patients (hereby referred to as potential NMOSD patients, pNMOSD) are diagnosed as myelitis, optic neuritis, chronic relapsing inflammatory optic neuritis or MS even though they might be seronegative NMOSD (Mealy et al., 2012). The absence of a biomarker makes their management more challenging. We hypothesized that retinal OCT imaging could be used to categorize these subjects as more similar to definite NMOSD than to MS.