A multivariate metabolic imaging marker for behavioral variant frontotemporal dementia

Area of Science:

• Neuroimaging
• Neurology
• Biomarker Discovery

Background:

• Behavioral variant frontotemporal dementia (bvFTD) presents with significant heterogeneity.
• Multivariate imaging biomarkers are crucial for understanding bvFTD.
• Current diagnostic methods require enhancement for this complex condition.

Purpose of the Study:

• To identify a novel multivariate imaging biomarker for bvFTD.
• To assess the diagnostic accuracy and clinical correlates of this biomarker.
• To establish a quantitative tool for bvFTD diagnosis and staging.

Main Methods:

• Principal component analysis (PCA) was applied to 18F-fluorodeoxyglucose-positron emission tomography (FDG-PET) scans from bvFTD and healthy subjects.
• A cohort of 148 dementia patients across three centers (Feinstein Institute, Technical University of Munich, ADNI) was studied.
• The identified pattern, bFDRP, was validated in independent testing sets for phenotypic specificity and clinical correlates.

Main Results:

• The bFDRP was identified and reproduced across independent cohorts, explaining significant voxel variance (up to 24.1%).
• Validation demonstrated high diagnostic accuracy (AUC = 0.97) for distinguishing bvFTD.
• bFDRP expression was significantly elevated in bvFTD patients and correlated with disease duration and severity.

Conclusions:

• The bFDRP serves as a reliable quantitative imaging marker for bvFTD.
• This biomarker can aid in the differential diagnosis of bvFTD.
• bFDRP facilitates gauging the underlying disease process in bvFTD.