Combined socio-behavioral evaluation improves the differential diagnosis between the behavioral variant of frontotemporal dementia and Alzheimer's disease: In search of neuropsychological markers

Background: Current diagnostic criteria for behavioral variant of frontotemporal dementia (bvFTD) and typical Alzheimer's disease (AD) include a differential pattern of neuropsychological impairments (episodic memory deficit in typical AD and dysexecutive syndrome in bvFTD). There is, however, large evidence of a frequent overlap in neuropsychological features, making the differential diagnosis extremely difficult. Objectives: In this retrospective study, we evaluated the diagnostic value of different cognitive and neurobehavioral markers in bvFTD and AD patient groups. Methods:We included 95 dementia patients with a clinical and biomarker evidence of bvFTD (n = 48) or typical AD (n = 47) pathology. A clinical 2-year follow-up confirmed clinical classification. Performances at basic cognitive tasks (memory, executive functions, visuo-spatial, language) as well as social cognition skills and neurobehavioral profiles have been recorded. A stepwise logistic regression model compared the neuropsychological profiles between groups and assessed the accuracy of cognitive and neurobehavioral markers in discriminating bvFTD from AD. Results: Statistical comparison between patient groups proved social cognition and episodic memory impairments as main cognitive signatures of bvFTD and AD neuropsychological profiles, respectively. Only half of bvFTD patients showed attentive/executive deficits, questioning their role as cognitive marker of bvFTD. Notably, the large majority of bvFTD sample (i.e., 70%) poorly performed at delayed recall tasks. Logistic regression analysis identified social cognition performances, Frontal Behavioral Inventory and Mini-Mental State Examination scores as the best combination in distinguishing bvFTD from AD. Conclusion: Social cognition tasks and socio-behavioral questionnaires are recommended in clinical settings to improve the accuracy of early diagnosis of bvFTD.