Six-Protein CSF Panel Accurately Identifies Frontotemporal Dementia Subtypes
A 2,900-protein CSF screen reveals distinct molecular signatures for genetic FTD subtypes and two diagnostic panels with >94% accuracy.
Summary
Frontotemporal lobar degeneration (FTLD) is a devastating early-onset dementia with no reliable fluid biomarkers. Researchers screened over 2,900 proteins in cerebrospinal fluid from genetic FTLD patients and healthy controls, identifying distinct protein signatures for each major genetic subtype. MAPT gene carriers showed 63 upregulated proteins tied to immune pathways, C9orf72 carriers had four proteins linked to energy metabolism, and GRN carriers displayed six proteins related to neuronal development. From these findings, the team built two diagnostic panels: a six-protein FTLD panel and a seven-protein TDP-pathology panel, both achieving AUC scores above 0.94 in a separate validation cohort. NEFL and TPM3 emerged as consistently significant across all comparisons. These panels could meaningfully improve early diagnosis and patient stratification for clinical trials targeting genetic forms of dementia.
Detailed Summary
Frontotemporal lobar degeneration (FTLD) is a leading cause of early-onset dementia, yet clinicians currently lack validated fluid biomarkers to diagnose it or distinguish its genetic subtypes. This gap delays diagnosis and complicates enrollment in targeted clinical trials, making the search for reliable cerebrospinal fluid (CSF) markers a high priority in neurodegeneration research.
Researchers from a large international consortium analyzed CSF from participants in two well-characterized cohorts. The discovery cohort, drawn from the GENFI study, included 47 symptomatic pathogenic variant carriers across C9orf72, GRN, MAPT, and TARDBP mutations, 124 presymptomatic carriers, and 57 healthy non-carriers. A validation cohort of 132 clinically diagnosed FTLD-spectrum patients and 32 controls was used for replication. Proximity extension assay technology enabled simultaneous measurement of more than 2,900 proteins per sample.
The analysis revealed 23 proteins significantly dysregulated in symptomatic carriers overall. Gene subgroup analysis uncovered strikingly different molecular landscapes: MAPT carriers showed 63 upregulated proteins enriched in immune function pathways, C9orf72 carriers had four proteins linked to energy metabolism, and GRN carriers showed six proteins tied to neuronal development and axonal projection. NEFL (neurofilament light chain) and TPM3 were consistently significant across all genetic groups and both cohorts, underscoring their robustness as disease markers.
Two diagnostic panels were developed using LASSO regression. The FTLD panel (NEFL, RBFOX3, NPTX1, TFF1, ENTPD5, CNP) achieved an AUC of 0.94 in the validation cohort. The TDP-pathology panel (NEFL, RBFOX3, CBLN4, ENTPD5, CCL25, CNP, MMP1) reached an AUC of 0.96, demonstrating excellent discriminative power for underlying TDP pathology specifically.
These findings lay important groundwork for clinical biomarker development in FTLD. Limitations include the abstract-only basis for this summary, relatively small subgroup sizes, and the need for further prospective validation before clinical adoption.
Key Findings
- A six-protein CSF panel distinguished FTLD from controls with an AUC of 0.94 in a validation cohort.
- A seven-protein panel specifically identified TDP-pathology FTLD with an AUC of 0.96.
- NEFL and TPM3 were consistently dysregulated across all genetic FTLD subtypes and both cohorts.
- MAPT carriers showed 63 upregulated immune-pathway proteins; C9orf72 showed energy metabolism changes; GRN showed neuronal development proteins.
- Presymptomatic carriers were included, opening a path toward earlier biomarker-based detection before symptom onset.
Methodology
The study used proximity extension assay technology to measure over 2,900 CSF proteins in a discovery cohort (228 participants from GENFI) and a separate validation cohort (164 participants). Linear regression adjusted for age and sex identified differentially abundant proteins, while LASSO regression was used to build parsimonious diagnostic panels.
Study Limitations
This summary is based on the abstract only, as the full text is not open access. Sample sizes within individual genetic subgroups are relatively small, which may limit the generalizability of subgroup-specific findings. Prospective longitudinal validation in independent, ethnically diverse cohorts is needed before clinical implementation.
Enjoyed this summary?
Get the latest longevity research delivered to your inbox every week.
Enter your email to subscribe: