Distinctive Brain Atrophy Progression Patterns Aid Classification of Frontotemporal Dementia Variants
Edoardo Spinelli1, Francesca Orlandi1, Silvia Basaia4, Francesco Costa2, Stefano Pisano5, Alma Ghirelli1, Elisa Canu6, Veronica Castelnovo7, Elisa Sibilla7, Ilaria Bottale1, Giordano Cecchetti8, Francesca Caso9, Giuseppe Magnani9, Paola Caroppo10, Sara Prioni10, Cristina Villa10, Lucio Tremolizzo11, Ildebrando Appollonio11, Federico Verde12, Nicola Ticozzi13, Vincenzo Silani13, Massimo Filippi3, Federica Agosta3
1Neuroimaging Research Unit, Division of Neuroscience; Neurology Unit; and Center for Alzheimer’s disease and Related Disorders (CARD), 2Neuroimaging Research Unit, Division of Neuroscience, 3Neuroimaging Research Unit, Division of Neuroscience; Neurology Unit; and Center for Alzheimer’s disease and Related Disorders (CARD); and Neurotech Hub, IRCCS San Raffaele Scientific Institute and Vita-Salute San Raffaele University, 4Neuroimaging Research Unit, Division of Neuroscience; Neurotech Hub, Vita-Salute San Raffaele University, 5Neuroimaging Research Unit, Division of Neuroscience, 6Neuroimaging Research Unit, Division of Neuroscience; Center for Alzheimer’s disease and related disorders (CARD); and Neurotech Hub, Vita-Salute San Raffaele University, 7Neuroimaging Research Unit, Division of Neuroscience; Center for Alzheimer’s disease and related disorders (CARD), 8Neuroimaging Research Unit, Division of Neuroscience; Neurology Unit; Center for Alzheimer’s disease and related disorders (CARD); and Neurophysiology Service, 9Neurology Unit; Center for Alzheimer’s disease and related disorders (CARD), IRCCS San Raffaele Scientific Institute, 10Unit of Neurology 5-Neuropathology, Fondazione IRCCS Istituto Neurologico Carlo Besta, 11Neurology Unit, "San Gerardo" Hospital and University of Milano-Bicocca, 12Department of Neurology and Laboratory of Neuroscience, 13Department of Neurology and Laboratory of Neuroscience; and "Dino Ferrari" Center, Department of Pathophysiology and Transplantation, Università degli Studi di Milano, IRCCS Istituto Auxologico Italiano
Objective:
To describe longitudinal structural MRI correlates of clinically overlapping frontotemporal dementia (FTD) variants, describing their patterns of cortical and subcortical atrophy and assessing their utility for differential diagnosis and prognosis.
Background:
The FTD spectrum encompasses a wide range of clinical phenotypes, with predominant behavioral and/or language presentations. Recently, a variant with predominant right temporal atrophy and clinical features straddling behavioral variant of FTD (bvFTD) and semantic variant of primary progressive aphasia (svPPA) has been described as semantic behavioural variant FTD (sbvFTD).
Design/Methods:
Our cohort included 71 patients with a clinical diagnosis of bvFTD (n=45), sbvFTD (n=11), or svPPA (n=15) and 37 healthy controls. Participants were followed-up for up to 24 months. Regional cortical thickness and subcortical volumes were assessed using linear mixed-effect models. Support vector machine (SVM) algorithms were employed to classify subjects based on baseline and longitudinal patterns of atrophy.
Results:
At baseline, patients with sbvFTD had an intermediate brain atrophy pattern between bvFTD and svPPA, with right-predominant temporal pole involvement associated with significant right frontal atrophy. Longitudinally, bvFTD patients progressed widely in bilateral cortical regions and basal ganglia, while svPPA continued steady progression within the temporal lobes, and sbvFTD showed progression in the left temporal and frontal lobes with limited further volume loss in the right hemisphere. Baseline cortical thickness values of frontal regions were predictors of subsequent functional decline in bvFTD and sbvFTD. A multiclass SVM model provided a good diagnostic classification accuracy, with similar results when using baseline data only (82%) and adding longitudinal data (83%).
Conclusions:
Our results singled out sbvFTD as a relatively distinct entity, with early involvement of extra-temporal cortical and subcortical regions. Our findings could help in the definition of machine learning aided diagnostic and prognostic protocols based on neuroimaging biomarkers in FTD variants.
Disclaimer: Abstracts were not reviewed by Neurology® and do not reflect the views of Neurology® editors or staff.