Discriminating behavioral variant frontotemporal dementia from Alzheimer's disease via the Frontal Assessment Battery

Background: This study aimed to assess the ability of the frontal assessment battery (FAB) to differentiate Alzheimer's disease (AD) from behavioral variant frontotemporal dementia (bvFTD), as well as to test whether its discriminatory accuracy varies as a function of disease stage. Methods: The present retrospective cohort included N = 595 participants-153 with biomarker-supported AD, 96 with probable bvFTD, and 346 healthy controls (HCs). All HCs scored within the normal range on the Montreal Cognitive Assessment. In patients, disease severity was staged according to Clinical Dementia Rating scores, retrospectively derived from the Mini-Mental State Examination. Receiver operating characteristic (ROC) analyses were performed to assess the ability of demographically adjusted FAB scores to differentiate both patient groups from HCs and AD from bvFTD patients, including severity-stratified analyses. Binary logistic regression models adjusted for age and sex examined the contribution of individual FAB subtests to diagnostic classification. Results: The FAB showed excellent accuracy in distinguishing patients from HCs. In contrast, its ability to differentiate AD from bvFTD was modest overall and depended on disease severity, with adequate discrimination emerging solely at moderate-to-severe stages. At the subtest level, Lexical fluency was the only subtests that significantly differentiated bvFTD from AD, particularly in patients at the MCI/questionable dementia and moderate-to-severe dementia stage. No significant differences emerged for the remaining subtests. Discussion: The FAB is an excellent screening test for executive dysfunction in both bvFTD and AD, but its contribution to the differential diagnosis between these two conditions emerges only at moderate-to-severe stages.