To develop an MRI biomarker for the diagnosis of Multiple System Atrophy of the cerebellar type (MSA-C).

MSA-C is a sporadic, adult-onset synucleinopathy with autonomic neuropathy and ataxia. Distinguishing MSA-C from other ataxias can be challenging. We hypothesized that rate of change of pons and middle cerebellar peduncle (MCP) diameters would facilitate accuracy and timeliness of diagnosis.

Exploratory cohort (88 MSA-C, 44 spinocerebellar ataxia, 13 Friedreich’s ataxia, 21 idiopathic/other ataxia) had baseline/longitudinal measurements (2002-2014) of anteroposterior (AP) mid-pons/transverse MCP diameters on conventional MRI. 73 Human Connectome Project subjects were sampled to derive normative data and pons diameter-volume correlations. Validation cohort (49 MSA-C, 13 MSA-P, 99 other ataxias, 79 Parkinson’s disease [PD], 200 other movement disorders) underwent similar baseline/longitudinal measurements (2015-2021).

Normative data: pons diameter 23.9±1.6 mm and MCP diameter 16.4±1.4 mm. Pons diameter-volume correlation r=0.94, p<0.0001.

Exploratory cohort:

Comparing MSA-C to other ataxias at first scan, pons diameter 19.3±2.6 mm vs. 20.7±2.6 mm, p=0.0006 and MCP 12.0±2.6 mm vs. 14.3±2.1 mm, p<0.0001.

Longitudinal rate of change (mean [SE]) using mixed-model regression in pons/MCP in MSA-C vs. other ataxias were markedly different: pons -0.85 [0.04] mm/year vs. -0.09 [0.02] mm/year; MCP -0.84 [0.05] mm/year vs. -0.08 [0.02] mm/year, both p<0.0001.

Pons/MCP measures were indistinguishable between Possible, Probable, and Definite MSA-C. 

AUC analysis in MSA-C with a threshold pons diameter decline -0.5 mm/year yielded sensitivity 0.92/specificity 0.87; while MCP diameter decline -0.4 mm/year, yielded sensitivity 0.85/specificity 0.79.

Validation cohort:

We replicated the findings. In non-ataxia movement disorders, pons diameter at first scan 22.5±1.5 and MCP diameter 17.1±1.2 mm were significantly different from MSA/other ataxias.

A rate of change of mid-pons AP diameter and MCP diameters ~0.8 mm/year is sensitive and specific for the diagnosis of MSA-C, is necessary and sufficient, and represents a novel phenotypic imaging biomarker for use in clinical trials.