Cortical free-water imaging in familial frontotemporal dementia associated with MAPT, GRN, and C9orf72 pathogenic variants.

Background: Familial frontotemporal lobar degeneration (FTLD) caused by pathogenic variants in C9orf72, GRN, and MAPT provides a unique framework for evaluating imaging markers of genotype-specific neurodegeneration. Conventional cortical mean diffusivity (cMD) is sensitive to microstructural injury but is influenced by extracellular free-water effects. We investigated whether cortical free water (cFW) provides a sensitive imaging readout of cortical microstructural alterations across the three major genetic forms of familial FTLD, and compared its spatial distribution and clinical relevance with cMD and free-water-corrected tissue mean diffusivity (MD-t). Methods: We analyzed data from 324 participants from the ARTFL-LEFFTDS Longitudinal Frontotemporal Lobar Degeneration (ALLFTD) study, including 199 carriers of pathogenic variants in C9orf72 (n = 85), GRN (n = 56), or MAPT (n = 58), spanning asymptomatic and symptomatic stages, and 125 non-carrier family members. Surface-based cortical maps of cMD, MD-t, and cFW were generated from diffusion MRI. Group differences between each genotype and non-carrier controls were assessed using general linear models adjusted for age and sex. Associations with disease severity, measured by the CDR Results: Across C9orf72, GRN, and MAPT carriers, cFW showed the most spatially extensive cortical abnormalities relative to non-carrier controls, whereas MD-t effects were consistently more circumscribed than conventional cMD. This pattern was observed across all three genotypes, with particularly widespread cFW elevations in C9orf72 and GRN carriers and more frontotemporal-predominant alterations in MAPT carriers. cFW also exhibited the broadest positive associations with FTLD-CDR and plasma NfL across genotypes, while MD-t associations were more regionally restricted. These findings suggest that extracellular free-water changes contribute substantially to diffusion abnormalities in familial FTLD and provide information complementary to tissue-restricted diffusivity. Conclusions: cFW is a sensitive cross-genotype imaging marker of cortical microstructural alterations and disease burden in familial FTLD associated with C9orf72, GRN, and MAPT pathogenic variants. Compared with conventional cMD and MD-t, cFW captures more spatially extensive disease-related abnormalities and shows broader cortical associations with clinical severity and plasma NfL. Longitudinal studies are needed to determine whether cFW improves prediction of disease progression and sensitivity to change in genetic FTLD trials.