Frontotemporal dementia remains one of the most diagnostically elusive neurodegenerative conditions, partly because its genetic architecture is complex and testing strategies vary widely across clinical centers. Understanding which patients are most likely to carry a pathogenic variant — and which genetic pathways dominate — has direct consequences for family counseling, clinical trial enrollment, and eventually targeted therapy.
This French multicenter study applied a tiered three-step screening protocol to 2,747 individuals with sporadic or familial frontotemporal lobar degeneration (FTLD). The workflow began with plasma progranulin measurement as a low-cost triage tool, followed by GRN gene sequencing when progranulin fell below a defined threshold, then C9orf72 repeat expansion testing, and finally a 14-gene targeted panel in a subset of 1,279 patients stratified by age at onset and family history. The overall diagnostic yield reached 12.2%, with GRN mutations (n=73) and C9orf72 expansions (n=200) together accounting for 81.7% of all positive findings. The panel's incremental yield beyond those two genes was modest at 4.8%. Critically, family history emerged as the dominant predictor of a positive result — individuals with a family history of FTLD were four to five times more likely to receive a genetic diagnosis (32.1%) than sporadic cases (7.2%). Clinical presentation subtype, by contrast, did not significantly differentiate genetic versus non-genetic cases.
These findings carry meaningful implications for how neurogenetics services are structured. The tiered approach, anchored by an inexpensive plasma biomarker, appears cost-efficient: the progranulin assay alone efficiently gates GRN testing without requiring upfront comprehensive panel sequencing for everyone. The outsized contribution of just two genes — GRN and C9orf72 — suggests that for many clinical settings, broad panel sequencing may add limited value unless family history is present or initial screening is negative. A key limitation is the cohort's French origin, which may not fully represent the genetic frequencies seen in other populations where, for instance, MAPT mutations are proportionally more common. The study is also observational and cross-sectional, precluding causal inference about phenotypic progression. Nonetheless, with nearly 2,800 probands, this represents one of the largest real-world FTLD genetic diagnostic series to date, offering a practical framework that other centers could adapt.