Temporal order of clinical, imaging, and biomarker changes in frontotemporal lobar degeneration-associated syndromes

Background: The temporal sequence of clinical, imaging, and biological changes in sporadic frontotemporal lobar degeneration (FTLD)-associated syndromes remains poorly characterized, and a comprehensive biomarker cascade model is lacking. Methods: We developed a data-driven biomarker cascade model in 489 patients across the FTLD spectrum (211 behaviorial variant frontotemporal dementia [bvFTD], 129 primary progressive aphasia [PPA], 71 corticobasal syndrome [CBS], 66 progressive supranuclear palsy [PSP], and 12 FTD associated with amyotrophic lateral sclerosis [FTD-ALS]; 1904 patient-visit observations). Plasma, magnetic resonance imaging (MRI), and clinical biomarkers were modeled using sigmoid trajectories fitted to covariate-adjusted longitudinal data. Results: Plasma glial fibrillary acidic protein departed from normality earliest, followed by Trail Making Test Part B (TMT-B), white matter lesion volume, and neurofilament light chain. Insular atrophy showed the steepest transition among MRI measures; clinical dementia rating dementia staging instrument plus National Alzheimer's Coordinating Center behavior and language domains sum of boxes declined most steeply overall. TMT-B inflected earliest in bvFTD, whereas insula atrophy dominated in PPA. Conclusions: This first data-driven temporal cascade of multimodal biomarkers in sporadic FTLD-associated syndromes offers a framework for disease staging and stage-specific clinical trial design.