Brainstem Substructure Atrophy in Late-Onset GM2-Gangliosidosis Imaging Using Automated Segmentation

• Published in: Cerebellum (London, England) Vol. 24; no. 2; p. 50
• Main Authors: Rowe, Olivia E., Rangaprakash, D., Eichler, Florian S., Schmahmann, Jeremy D., Barry, Robert L., Stephen, Christopher D.
• Format: Journal Article
• Language: English
• Published: New York Springer US 18.02.2025; Springer Nature B.V
• Subjects: Adult; Aged; Ataxia; Atrophy; Atrophy - diagnostic imaging; Biomedical and Life Sciences; Biomedicine; Brain stem; Brain Stem - diagnostic imaging; Brain Stem - pathology; Brief Report; Cerebellum; Female; Gangliosidoses, GM2 - diagnostic imaging; Gangliosidoses, GM2 - pathology; Gangliosidosis; Humans; Image processing; Magnetic Resonance Imaging - methods; Male; Medulla oblongata; Mesencephalon; Middle Aged; Neurobiology; Neuroimaging; Neurology; Neurosciences; Pons; Segmentation; Superior cerebellar peduncle; Structural MRI; Tay-Sachs disease; Late-Onset GM2-Gangliosidosis; Sandhoff disease
• ISSN: 1473-4230; 1473-4222; 1473-4230
• DOI: 10.1007/s12311-025-01803-4

Late-Onset GM2-Gangliosidoses (LOGG) are rare, neurodegenerative lysosomal disorders that include late-onset Tay-Sachs (LOTS) and Sandhoff disease (LOSD) subtypes. Cerebellar atrophy is common, even in the absence of clinical ataxia, particularly in LOTS. Recent reports have also described brainstem atrophy in LOTS. We assessed brainstem substructure atrophy in LOGG, including LOSD. 10 LOGG patients (7 LOTS, 3 LOSD) and 7 age-matched controls had structural MRI brain imaging. A FreeSurfer brainstem substructure module was used for automatic segmentation and included the pons, medulla, superior cerebellar peduncle (SCP), midbrain, and total brainstem. Clinical ataxia severity was assessed with the LOTS Severity Scale, Brief Ataxia Rating Scale, Friedreich’s Ataxia Rating Scale and Scale for the Assessment and Rating of Ataxia. There were differences between LOGG and controls in the pons (12,785.06 ± 1,603.84 vs. 15,457.14 ± 2,748.41 mm 3 , p  = 0.0069) and SCP (196.93 ± 31.20 vs. 293.57 ± .70.16 mm 3 , p  = 0.0003). In LOTS vs. controls, there was similar pons ( p  = 0.0055) and SCP atrophy ( p  = 0.0023). The LOSD group was too small for independent comparisons. There were no significant associations between SCP/pons volume and clinical scales or disease duration. Cerebellar volume, which was analyzed in a previous study by Rowe et al. (2021), was relatively preserved in LOSD compared to the SCP/pons, while in LOTS, the pontocerebellar atrophy profile was dominated by cerebellar atrophy. These findings provide anatomical and clinical insights to the cerebellar/brainstem atrophy observed in LOGG and highlight a need to stratify LOGG by subtypes.