MRI Features in a Myelin Oligodendrocyte Glycoprotein Positive Cohort Using Current Diagnostic Criteria
Jodie M. Burton1, Fiona Costello2, Saerom Youn5, Abdullah Al-Ani3, Zarina Assis4, Napo Kasirye-Mbugua4
1Department of Clinical Neurosciences, Department of Community Health Sciences, 2Department of Clinical Neurosciences, Department of Surgery (Ophthalmology), 3Department of Surgery (Ophthalmology), 4Department of Imaging, University of Calgary, 5Department of Surgery (Ophthalmology), McMaster University
Objective:

Our objective was to examine MRI features and diagnoses in myelin oligodendrocyte glycoprotein (MOG) IgG1 antibody positive patients in Alberta.

Background:

In MOG antibody-associated disease (MOGAD), specific optic nerve, brain, and spinal cord MRI features can help distinguish MOGAD from such demyelinating diseases as neuromyelitis optica spectrum disorder and multiple sclerosis (MS).

Design/Methods:

In Alberta, MOG antibody testing uses a fixed cell-based assay. A data pull of all MOG IgG1 assays was undertaken from 07/2017 to 07/2023. Chart review was performed in positive patients, categorized by antibody titer, MRI category (ADEM, non-specific optic neuritis (ON), MOG-ON, transverse myelitis (TM), longitudinally extensive TM (LETM), MS-brain +/- cord, MOG-brain, and “other” brain), and diagnosis. Comparisons using WRST were performed for MRIs with and without MOGAD features.

Results:

Of 3831 tested patients, 208 were MOG positive, 158/208 with both a MOGAD-like presentation and MRI brain +/or orbits +/or spine. Ninety-five of 158 patients were diagnosed with MOGAD (80/95 with titers => 1/100 endpoint dilution and 15/95 with 1/10 endpoint dilution with supporting criteria). The most common alternative diagnosis was MS (28/158). Patients with MRI MOG-ON vs MRI-ON were younger (31.2y vs 40.1y, p=0.062), as were LETM vs TM patients (24.6y vs 46.5y, p=0.0001), and MOG-brain vs MS-brain patients (13.4y vs 36.9y, p=0.0006). MOGAD was diagnosed in 86-91% of patients with any demyelinating ON lesions owing to supportive clinical features, but more common in LETM vs TM (93% vs 11%), and MOG-brain vs MS-brain (100% vs 16%). Eighty percent of LETM patients had endpoint dilution => 1:100 vs 22% in TM, 50% vs 34% in MOG-brain vs MS-brain patients.  

Conclusions:

MOGAD MRI brain/cord features correlated with acceptable MOG antibody titers and MOGAD diagnosis, while both radiological and clinical ON features more often supported a MOGAD diagnosis, buoying current diagnostic criteria. Care is needed to avoid confusion with MS.

10.1212/WNL.0000000000205795