Molecular stratification of frontotemporal dementia through multi-omics integration and machine learning

Frontotemporal dementia (FTD) comprises clinically diverse behavioral, language, and motor-associated syndromes, while frontotemporal lobar degeneration (FTLD) refers to the underlying neuropathological spectrum. This distinction matters because similar clinical presentations can reflect different proteinopathies, and TDP-43 pathology also occurs in conditions such as limbic-predominant age-related TDP-43 encephalopathy. We critically review the evidence for molecular stratification across clinical phenotype, genetic etiology, neuropathological subtype, and disease stage, focusing on findings from longitudinal cohorts and genomic, transcriptomic, proteomic, and imaging studies. Studies in genetic FTD support serum neurofilament light chain as a marker of neuroaxonal injury and proximity to symptomatic conversion, although it does not identify the underlying proteinopathy. Cerebrospinal fluid (CSF) proteomic panels have shown encouraging discrimination of clinically defined FTD from controls or Alzheimer disease in independent cohorts, but reproducible differentiation of FTLD-tau from FTLD-TDP remains unresolved. Genetic and proteomic studies have also identified subtype-associated lysosomal, synaptic, RNA-processing, and inflammatory signatures, whose clinical specificity and prospective utility require further study. Machine learning can support feature selection and multimodal analysis, but matched evidence is limited that comprehensive multi-omics models outperform simpler biomarkers or clinical baselines in pathology-confirmed cohorts. We propose an evidence-based approach in which the intended use, biological compartment, disease stage, reference diagnosis, and external validation are specified before molecular stratification claims are made. This approach can guide the design of biomarker studies and genetically informed trials without treating molecular association, computational prediction, or target engagement as proof of pathological classification or treatment efficacy.

Authors

Institutions

Publication Details

Journal
Molecular Medicine
Published
2026-11-21
DOI
https://doi.org/10.1186/s10020-026-01701-3
Primary Topic
Dementia and Cognitive Impairment Research
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Molecular stratification of frontotemporal dementia through multi-omics integration and machine learning

Faez Iqbal Khan, Mohammed Ageeli Hakami
Molecular Medicine
Dementia and Cognitive Impairment Research
article

Molecular stratification of frontotemporal dementia through multi-omics integration and machine learning

Faez Iqbal Khan, Mohammed Ageeli Hakami
article en

Abstract

Frontotemporal dementia (FTD) comprises clinically diverse behavioral, language, and motor-associated syndromes, while frontotemporal lobar degeneration (FTLD) refers to the underlying neuropathological spectrum. This distinction matters because similar clinical presentations can reflect different proteinopathies, and TDP-43 pathology also occurs in conditions such as limbic-predominant age-related TDP-43 encephalopathy. We critically review the evidence for molecular stratification across clinical phenotype, genetic etiology, neuropathological subtype, and disease stage, focusing on findings from longitudinal cohorts and genomic, transcriptomic, proteomic, and imaging studies. Studies in genetic FTD support serum neurofilament light chain as a marker of neuroaxonal injury and proximity to symptomatic conversion, although it does not identify the underlying proteinopathy. Cerebrospinal fluid (CSF) proteomic panels have shown encouraging discrimination of clinically defined FTD from controls or Alzheimer disease in independent cohorts, but reproducible differentiation of FTLD-tau from FTLD-TDP remains unresolved. Genetic and proteomic studies have also identified subtype-associated lysosomal, synaptic, RNA-processing, and inflammatory signatures, whose clinical specificity and prospective utility require further study. Machine learning can support feature selection and multimodal analysis, but matched evidence is limited that comprehensive multi-omics models outperform simpler biomarkers or clinical baselines in pathology-confirmed cohorts. We propose an evidence-based approach in which the intended use, biological compartment, disease stage, reference diagnosis, and external validation are specified before molecular stratification claims are made. This approach can guide the design of biomarker studies and genetically informed trials without treating molecular association, computational prediction, or target engagement as proof of pathological classification or treatment efficacy.

Molecular Medicine
Shaqra University (SA), King Salman Center for Disability Research (SA), Xi’an Jiaotong-Liverpool University (CN)
King Salman Center for Disability Research
Good health and well-being
Dementia and Cognitive Impairment Research
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon