Neurofilaments and progranulin are related to atrophy in frontotemporal lobar degeneration - A transdiagnostic study cross-validating atrophy and fluid biomarkers

Abstract

Introduction: Frontotemporal lobar degeneration (FTLD) encompasses behavioral variant frontotemporal dementia (bvFTD), progressive supranuclear palsy, corticobasal syndrome/degeneration, and primary progressive aphasias (PPAs). We cross-validated fluid biomarkers and neuroimaging.

Methods: Seven fluid biomarkers from cerebrospinal fluid and serum were related to atrophy in 428 participants including these FTLD subtypes, logopenic variant PPA (lvPPA), Alzheimer's disease (AD), and healthy subjects. Atrophy was assessed by structural magnetic resonance imaging and atlas-based volumetry.

Results: FTLD subtypes, lvPPA, and AD showed specific profiles for neurofilament light chain, phosphorylated heavy chain, tau, phospho-tau, amyloid beta_1-42 from serum/cerebrospinal fluid, and brain atrophy. Neurofilaments related to regional atrophy in bvFTD, whereas progranulin was associated with atrophy in semantic variant PPA. Ubiquitin showed no effects.

Discussion: Results specify biomarker and atrophy patterns in FTLD and AD supporting differential diagnosis. They identify neurofilaments and progranulin in interaction with structural imaging as promising candidates for monitoring disease progression and therapy.

Highlights: Study cross-validated neuroimaging and fluid biomarkers in dementia. Five kinds of frontotemporal lobar degeneration and two variants of Alzheimer's disease. Study identifies disease-specific fluid biomarker and atrophy profiles. Fluid biomarkers and atrophy interact in a disease-specific way. Neurofilaments and progranulin are proposed as biomarkers for diagnosis and therapy.

Keywords: amyloid; atrophy; fluid biomarkers; frontotemporal dementia; frontotemporal lobar degeneration; magnetic resonance imaging; neurofilaments; phospho‐tau; progranulin; tau; ubiquitin.

FIGURE 1

Fluid biomarker concentrations in healthy participants and AD, bvFTD, nfvPPA, svPPA, lvPPA, PSP, and CBS. Red asterisks denote significant differences between serum NfL levels of patients and healthy participants. Black bars and asterisks show significant differences of biomarker levels between patient cohorts. Significances: *p < .05; **p < .01; ***p < .001; ****p < .0001. Means ± standard deviation. Blue dashed lines indicate thresholds for pathological values provided by the central laboratory. AD, Alzheimer's disease; bvFTD, behavioral variant frontotemporal dementia; CBS, corticobasal syndrome; CSF, cerebrospinal fluid; lvPPA, logopenic variant primary progressive aphasia; n, number of available participants for the respective measure; nfvPPA, non‐fluent variant primary progressive aphasia; PSP, progressive supranuclear palsy; svPPA, semantic variant primary progressive aphasia.

FIGURE 2

Brain volume differences between patients and controls. Bonferroni corrected results of permutation ANCOVA with age as covariate, testing for volume differences between each patient cohort, AD, lvPPA, bvFTD, nfvPPA, svPPA, PSP, CBS, and healthy controls, separately for all anatomical structures. Only significantly differing areas indicating atrophy are displayed with color coding illustrating effect strength in terms of Hedge's g. Yellow and white colors represent a strong effect and red a weaker effect. Note that the range of effect strengths varies between cohorts. N denotes the number of participants. Significance: p < .05. Left side is left. AD, Alzheimer's disease; bvFTD, behavioral variant frontotemporal dementia; CBS, corticobasal syndrome; lvPPA, logopenic variant primary progressive aphasia; nfvPPA, non‐fluent variant primary progressive aphasia; PSP, progressive supranuclear palsy; svPPA, semantic variant primary progressive aphasia.

FIGURE 3

Whole brain, as well as total gray and white matter volumes in healthy participants and AD, bvFTD, nfvPPA, svPPA, lvPPA, PSP, and CBS. Means ± standard deviation. Total brain volume refers to gray and white matter volumes without cerebrospinal fluid. AD, Alzheimer's disease; bvFTD, behavioral variant frontotemporal dementia; CBS, corticobasal syndrome; lvPPA, logopenic variant primary progressive aphasia; n, number of available participants for respective measure; nfvPPA, non‐fluent variant primary progressive aphasia; PSP, progressive supranuclear palsy; svPPA, semantic variant primary progressive aphasia.

FIGURE 4

Correlations between fluid biomarkers and brain volumetry. Significant results of partial Spearman correlation between biomarkers and brain volumes, separately for each patient cohort (respective left panel). Significant volume differences between patient cohorts and healthy controls indicating atrophy (respective middle panel). Overlap between correlation and atrophy results (respective right panel). Results are Bonferroni corrected and displayed with color coding indicating effect strength in terms of Spearman's ρ in green and Hedge's g in red; overlap is indicated by blue. Bright yellow or white color represents a strong effect, green/red a weaker effect. Note that the range of effect strengths varies between cohorts. Significance level p < .05. Left side is left. AD, Alzheimer's disease; bvFTD, behavioral variant frontotemporal dementia; CSF, cerebrospinal fluid; NfL, neurofilament light chain; pNfH, phosphorylated neurofilament heavy chain; PSP, progressive supranuclear palsy; svPPA, semantic variant primary progressive aphasia.

FIGURE 5

Partial Spearman correlations between whole brain and white matter volumes and NfL (serum and cerebrospinal fluid) concentrations across all disease groups. Total brain volume refers to gray and white matter volumes without cerebrospinal fluid. Rho is the age‐corrected correlation coefficient Spearman's ρ, Significance level p < .05. For disease‐specific results in bvFTD and AD, see Table S18. AD, Alzheimer's disease; bvFTD, behavioral variant frontotemporal dementia; CBS, corticobasal syndrome; CSF, cerebrospinal fluid; lvPPA, logopenic variant primary progressive aphasia; N, number of available participants for the respective measure; NfL, neurofilament light chain; nfvPPA, non‐fluent variant primary progressive aphasia; PSP, progressive supranuclear palsy; svPPA, semantic variant primary progressive aphasia.