This study applied SFC to test the hypothesis that topological stepwise architecture propagating from the disease epicenter would shape patterns of grey matter (GM) atrophy in a cohort of patients with progressive supranuclear palsy (PSP).

MRI connectomics is an ideal tool to test the model of network-based spread of pathological protein aggregates in neurodegenerative disorders. Stepwise functional connectivity (SFC) is a graph-theory-based neuroimaging method, which detects whole-brain functional couplings of a selected region of interest, at increasing link-step topological distances.

Twenty-eight patients with PSP and 50 healthy controls underwent brain magnetic resonance imaging (MRI) on a 3T scanner, including 3D-T1 weighted and resting-state functional MRI sequences. GM was parcellated into 90 regions using the Automated Anatomical Labeling (AAL) atlas. Correlations between SFC architecture in controls and atrophy patterns in PSP patients were tested. The disease epicenter was identified as the peak of atrophy observed in an independent cohort of 13 cases with post mortem confirmation of PSP pathology, and used as seed region for SFC analysis.

The disease epicenter was identified in the left midbrain tegmental region. Compared with controls, PSP patients showed prevalent atrophy in the subcortical GM (mostly, in the thalami and caudate nuclei), but also in frontal, parietal and cerebellar cortical regions. For each region of the AAL atlas, a strong correlation was found between average link-step distance from the left midbrain in controls and mean normalized GM volume in PSP patients (r=0.37, p<0.001).