Frontotemporal dementia treatment: an umbrella review of meta-analyses of randomized controlled trials
Year: 2026
Type: article
Abstract: Frontotemporal dementia (FTD) has no approved disease-modifying therapies, and the role and the evidence of symptomatic treatment are still debated. We conducted an umbrella review of meta-analyses of randomized controlled trials (RCTs) to summarise treatment efficacy. Three databases were searched from inception to 1 November 2025. We included systematic reviews with meta-analyses of randomised controlled trials (RCTs) in people with FTD, comparing any active intervention with placebo, sham, no treatment or an active control, published in English. Eligible outcomes were any clinical outcome, including behavioural and psychological symptoms of dementia, cognitive and language function, caregiver burden and activities of daily living; in practice, the outcomes reported by the included meta-analyses were predominantly cognitive and language tests. Network meta-analyses were excluded a priori. Two reviewers screened the literature and extracted data. Overlap of primary RCTs across the included reviews was quantified using the corrected covered area (CCA). Methodological quality was assessed with AMSTAR 2 and certainty of evidence with GRADE. Of 879 records identified, 558 were screened after removal of duplicates and 29 were assessed in full text; five meta-analyses were included, covering 22 unique RCTs and 15 outcomes. The outcomes were mainly cognitive and language tests. Meta-analysis quality was high in four and moderate in one. High-certainty evidence indicated that memantine, the only pharmacological treatment for which a meta-analysis of FTD-specific RCTs was available, was not superior to placebo for Global Clinical Impression, MMSE and Neuropsychiatric Inventory scores, or caregiver burden. In PPA, non-invasive brain stimulation (NIBS) considered overall showed no benefit for naming (SMD 0.06, 95% CI − 0.05 to 0.17) and a very modest improvement in phonemic fluency (SMD 0.11, 95% CI 0.02 to 0.21), both with very low certainty. In tDCS-specific analyses, tDCS improved trained naming (SMD 0.67, 95% CI 0.23 to 1.11; low certainty) and untrained naming (SMD 0.31, 95% CI 0.15 to 0.47; moderate certainty), with no consistent effects on other outcomes. Meta-analytic evidence for the symptomatic treatment of FTD is scarce and concentrated on a small number of interventions and outcomes. High-certainty evidence did not support a benefit of memantine over placebo for the evaluated outcomes. Small trials suggested a possible benefit of tDCS on naming in PPA; this signal warrants further investigation as an adjunctive intervention rather than clinical adoption. Larger phenotype-stratified trials with patient- and caregiver-relevant endpoints are needed.
Institutions: Karolinska Institutet, Azienda Socio Sanitaria Territoriale degli Spedali Civili di Brescia, University of Brescia, Bezmiâlem Vakıf Üniversitesi, Biruni University, Anglia Ruskin University, University of Ottawa, Ottawa Hospital, Ottawa Hospital Research Institute, Charité – Universitätsmedizin Berlin, IRCCS San Camillo Hospital, Griffith University, University of Primorska, University of Palermo, Saint Camillus International University of Health and Medical Sciences
Language: English
Subfield: Psychiatry and Mental health
Field: Medicine
Domain: Health Sciences
Keywords: frontotemporal dementia, symptomatic treatment, disease-modifying therapy, memantine, non-invasive brain stimulation, transcranial direct current stimulation, primary progressive aphasia, naming, phonemic fluency, behavioral and psychological symptoms of dementia, caregiver burden, treatment efficacy, umbrella review of meta-analyses, methodological quality of meta-analyses, quality of evidence
Open Access status: hybrid
Estimated APC paid: $4,490
OpenAlex ID: W7236744045
