Neuromyelitis Optica Spectrum Disorder: Therapeutic Innovations and Complex Decision‐Making

Abstract

The rare coincidence of severe optic nerve and spinal cord damage was first described in the early 19th century. Devic from Lyon, prompted by a patient of his own, in 1894 summarized and analyzed these early observations and defined a disease that was to carry his name. For many years it was considered a variant of multiple sclerosis. Differences in clinical, magnetic resonance imaging (MRI), and cerebrospinal fluid findings identified in the 1990s paved the way for a change in concept. The discovery by the Mayo group in 2004 of a pathogenic autoantibody against the aquaporin-4 water channel localized in endfeet of astrocytes established the syndrome as a nosological entity in its own right. The detection of this pathognomonic antibody allowed investigators to include presentations involving other areas of the central nervous system under the rubric of neuromyelitis optica spectrum disorder (NMOSD). The disease is typically characterized by relapses that can produce marked disability. Exacerbations can occur even after long periods of remission and mandate continuous preventive therapy. A large body of evidence implicates aberrant cellular and prominently humoral immunity involving T cells, B cells, plasmablasts, aquaporin-4 immunoglobulin (IgG), proinflammatory cytokines, B-cell differentiation and survival factors, and complement activation products in the pathogenesis of NMOSD. Recognition of their pathogenetic roles has enabled the design of innovative and more specific immunointerventions beyond the widely used classical immunosuppressants such as azathioprine and mycophenolate mofetil. In astounding synchrony, double-blind randomized placebo-controlled phase 2/3 trials were conducted targeting CD19-positive B cells including precursors and plasmablasts, blocking terminal complement activation, or disrupting interleukin-6 signaling pathways. All approaches proved to be efficacious in markedly reducing NMOSD relapses. The demonstration in NMO brains of deposition of both immunoglobulins and C9a suggested formation of terminal complement complexes that damaged astrocytes and neurons. These findings encouraged exploration of the potential utility of the complement blocker eculizumab. This humanized monoclonal antibody inhibits C5 convertase and subsequently generation of the proinflammatory chemotactic peptide C5a and prevents assembly of the lytic terminal complement complex C5b-9. It had been developed for use in paroxysmal nocturnal hemoglobinuria and hemolytic uremic syndrome and is also licensed for myasthenia gravis. The phase 3 Prevention of relapses in Neuromyelitis optica (PREVENT) trial randomized 143 aquaporin-4 IgG-positive adults to either intravenous (iv) eculizumab or placebo in the presence (76%) or absence of continued immunosuppressive treatments (azathioprine, mycophenolate, glucocorticosteroids). In a time-to-event design, the primary outcome was the first relapse and secondary outcomes were annualized relapse rates, Expanded Disability Status Scale (EDSS) changes, and quality of life measures. Comparative analysis, after 23 of 24 prespecified relapses had occurred, showed superiority of eculizumab over placebo. Three percent of participants randomized to eculizumab and 43% of those who received placebo had an adjudicated relapse, mostly myelitis. The median time to first relapse was 103 weeks for the placebo group but could not be established in the eculizumab group because of insufficient numbers of relapses. The annualized relapse rate went down from 0.35 in the placebo group to 0.02 in the eculizumab group. Adverse events including upper respiratory infections and headaches were more frequent in the eculizumab group. One patient in the eculizumab group who was also on azathioprine and had a history of lung disease died from pulmonary empyema. All of the participants had received meningococcal vaccination prior to trial entry, and none had a meningococcal infection.