Joao Vitor Mahler1, Fiona Salas1, Mulan Jiang1, James Nguyen1, Rebecca Salky1, Gabriela Romanow1, Marina Vilardo1, Monique Anderson1, Natalia Drosu1, Takahisa Mikami1, Rebecca Gillani1, Anastasia Vishnevetsky1, Philippe-Antoine Bilodeau1, Michael Levy1, Marcelo Matiello1
1Division of Neuroimmunology, Department of Neurology, Massa General Brigham, Harvard Medical School
Objective:
To determine the frequency of initial negative followed by positive anti-MOG IgG1 cell-based assays (CBAs) in patients with myelin oligodendrocyte glycoprotein antibody–associated disease (MOGAD).
Background:
MOGAD is serologically identified by CBAs for anti-MOG IgG1. While the occurrence of false positives for MOG has been well described, less is known about false negatives or post-presentation seroconversion. Here, we describe patients who were initially anti-MOG IgG1 negative at the acute phase and later were found to be positive.
Design/Methods:
We queried the MassGeneral Neuroimmunology database (2016–2025), identifying patients with ≥1 positive anti-MOG IgG test. Inclusion criteria for this case series were fulfillment of the MOGAD-2023 criteria, and having an initial anti-MOG IgG1 CBA that was negative followed by a positive test on repeat serology.
Results:
Four patients (1.1%) out of 355, with a median age at onset of 50.7 years (IQR 37.3–65.8) were identified; all were male and fulfilled MOGAD-2023 criteria. Index phenotypes consisted of unilateral severe ON (n=1), bilateral ON (n=1), LETM (n=1), and meningoencephalitis/ADEM (n=1). Oligoclonal bands were positive in 1 of 3 patients who were tested (5 unique CSF-bands). First negative serology was obtained a median of 4 days [1–7] from onset. Positive serology occurred at a median of 74 days [38–242] post-onset; first-positive titers ranged 1:20–1:1000. No chronic immunotherapy or plasma exchange was given before the first test; one had received rituximab when the test returned positive. Over a median of 4.8 months [2.3–18.6] of follow-up, two patients had relapses after seroconversion.
Conclusions:
Our findings might be explained by analytical issues (assay sensitivity, prozone effect) or by disease evolution (antibody-epitope maturation or fluctuating titers). Isotype mismatch may also reduce detection, as MOG-reactive IgA and IgG3 have been reported. These considerations support repeat testing with serial dilutions, when clinicoradiologic suspicion is high. The retrospective design, and short follow-up limit inference.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.