Abstract
BACKGROUND: Myelin oligodendrocyte glycoprotein antibody-associated encephalomyelitis (MOG-EM; also termed MOG antibody-associated disease, MOGAD) is the most important differential diagnosis of both multiple sclerosis and neuromyelitis optica spectrum disorders. A recent proposal for new diagnostic criteria for MOG-EM/MOGAD explicitly recommends the use of immunoglobulin G subclass 1 (IgG1)- or IgG crystallizable fragment (Fc) region-specific assays and allows the use of heavy-and-light-chain-(H+L) specific assays for detecting MOG-IgG. By contrast, the utility of MOG-IgG3-specific testing has not been systematically evaluated. OBJECTIVE: To assess whether the use of MOG-IgG3-specific testing can improve the sensitivity of MOG-IgG testing. METHODS: Re-testing of 22 patients with a definite diagnosis of MOG-EM/MOGAD and clearly positive MOG-IgG status initially but negative or equivocal results in H+L- or Fc-specific routine assays later in the disease course (i.e. patients with spontaneous or treatment-driven seroreversion). RESULTS: In accordance with previous studies that had used MOG-IgG1-specific assays, IgG subclass-specific testing yielded a higher sensitivity than testing by non-subclass-specific assays. Using subclass-specific secondary antibodies, 26/27 supposedly seroreverted samples were still clearly positive for MOG-IgG, with MOG-IgG1 being the most frequently detected subclass (25/27 [93%] samples). However, also MOG-IgG3 was detected in 14/27 (52%) samples (from 12/22 [55%] patients). Most strikingly, MOG-IgG3 was the predominant subclass in 8/27 (30%) samples (from 7/22 [32%] patients), with no unequivocal MOG-IgG1 signal in 2 and only a very weak concomitant MOG-IgG1 signal in the other six samples. By contrast, no significant MOG-IgG3 reactivity was seen in 60 control samples (from 42 healthy individuals and 18 patients with MS). Of note, MOG-IgG3 was also detected in the only patient in our cohort previously diagnosed with MOG-IgA(+)/IgG(-) MOG-EM/MOGAD, a recently described new disease subvariant. MOG-IgA and MOG-IgM were negative in all other patients tested. CONCLUSIONS: In some patients with MOG-EM/MOGAD, MOG-IgG is either exclusively or predominantly MOG-IgG3. Thus, the use of IgG1-specific assays might only partly overcome the current limitations of MOG-IgG testing and-just like H+L- and Fcγ-specific testing-might overlook some genuinely seropositive patients. This would have potentially significant consequences for the management of patients with MOG-EM/MOGAD. Given that IgG3 chiefly detects proteins and is a strong activator of complement and other effector mechanisms, MOG-IgG3 may be involved in the immunopathogenesis of MOG-EM/MOGAD. Studies on the frequency and dynamics as well as the clinical and therapeutic significance of MOG-IgG3 seropositivity are warranted.