Plasma Tau and Neurofilament Light in Frontotemporal Lobar Degeneration and Alzheimer Disease

Abstract

Objective: To test the hypothesis that plasma total tau (t-tau) and neurofilament light chain (NfL) concentrations may have a differential role in the study of frontotemporal lobar degeneration syndromes (FTLD-S) and clinically diagnosed Alzheimer disease syndromes (AD-S), we determined their diagnostic and prognostic value in FTLD-S and AD-S and their sensitivity to pathologic diagnoses.

Methods: We measured plasma t-tau and NfL with the Simoa platform in 265 participants: 167 FTLD-S, 43 AD-S, and 55 healthy controls (HC), including 82 pathology-proven cases (50 FTLD-tau, 18 FTLD-TDP, 2 FTLD-FUS, and 12 AD) and 98 participants with amyloid PET. We compared cross-sectional and longitudinal biomarker concentrations between groups, their correlation with clinical measures of disease severity, progression, and survival, and cortical thickness.

Results: Plasma NfL, but not plasma t-tau, discriminated FTLD-S from HC and AD-S from HC. Both plasma NfL and t-tau were poor discriminators between FLTD-S and AD-S. In pathology-confirmed cases, plasma NfL was higher in FTLD than AD and in FTLD-TDP compared to FTLD-tau, after accounting for age and disease severity. Plasma NfL, but not plasma t-tau, predicted clinical decline and survival and correlated with regional cortical thickness in both FTLD-S and AD-S. The combination of plasma NfL with plasma t-tau did not outperform plasma NfL alone.

Conclusion: Plasma NfL is superior to plasma t-tau for the diagnosis and prediction of clinical progression of FTLD-S and AD-S.

Classification of evidence: This study provides Class III evidence that plasma NfL has superior diagnostic and prognostic performance vs plasma t-tau in FTLD and AD.

Figure 1. Group Differences in Plasma Total Tau (t-Tau) and Neurofilament Light (NfL) Concentrations

Group differences in the plasma levels of t-tau (A) and NfL (B) between the main clinical groups. Group differences in the plasma levels of t-tau (C) and NfL (D) between frontotemporal lobar degeneration syndromes (FTLD-S) subgroups, the Alzheimer disease syndromes (AD-S) group, and healthy controls. Differences in the plasma levels of t-tau (E) and NfL (F) between major neuropathologic subtypes. In (E, F), participants with C9orf72 (n = 11) or GRN (n = 7) mutations were included in the FTLD-TDP group (n = 27), while participants with a MAPT mutation (n = 4) were included in the FTLD-tau group (n = 52). The Alzheimer disease (AD) group in (E, F) included all AD-S with pathologic confirmation of AD or a positive amyloid PET (n = 30). *p < 0.001, Bonferroni post hoc test. a: Inferior to all other groups (p < 0.05, Bonferroni post hoc test) expect nonfluent/agrammatic variant of primary progressive aphasia (nfvPPA) (p = 0.08). b: Superior to all other groups (p < 0.05, Bonferroni post hoc test). c: Inferior to all other groups (p < 0.05, Bonferroni post hoc test). ns: no statistically significant differences between groups (p > 0.05). bvFTD = behavioral variant frontotemporal dementia; CBS = corticobasal syndrome; FTD-ALS = frontotemporal dementia with amyotrophic lateral sclerosis; FTLD = frontotemporal lobar degeneration; HC = healthy control; PSP = progressive supranuclear palsy; svPPA = semantic variant of primary progressive aphasia.

Figure 2. Diagnostic Value of Plasma Total Tau (t-Tau) and Neurofilament Light (NfL) for the Differentiation of Frontotemporal Lobar Degeneration Syndromes (FTLD-S), Alzheimer Disease Syndromes (AD-S), and Healthy Controls (HC)

Diagnostic value of plasma t-tau (A) and NfL (B) for the differentiation of FTLD-S, AD-S, and HC. AUC = area under the curve.

Figure 3. Estimates of Annualized Clinical Deterioration as a Function of Baseline Plasma Biomarkers in Frontotemporal Lobar Degeneration Syndromes (FTLD-S) and Alzheimer Disease Syndromes (AD-S)

Clinical Dementia Rating plus National Alzheimer's Coordinating Center FTLD sum of boxes (CDR+NACC/FTLD-SB) estimates were obtained from linear mixed-effects models adjusted for age, sex, and basal CDR+NACC/FTLD-SB. For illustrative purposes, we show the groups with high levels of plasma biomarker (higher than the median) and low levels of plasma biomarker (lower than the median). Error bars represent 95% confidence intervals. NfL = neurofilament light.

Figure 4. Relationship Between Plasma Biomarkers and Cortical Thickness in Frontotemporal Lobar Degeneration Syndromes (FTLD-S) and Alzheimer Disease Syndromes (AD-S) Groups

Group comparison of cortical thickness between healthy controls (HC) and FTLD-S (A) and AD-S (B). Correlation between basal plasma levels of tau and cortical thickness in FTLD-S group (C); correlation between basal plasma levels of total tau (t-tau) and cortical thickness in AD-S group (D); correlation between basal plasma levels of NfL and cortical thickness in FTLD-S group (E); correlation between basal plasma levels of NfL and cortical thickness in AD-S group (F). For group comparisons, only clusters that survived false discovery rate correction (p < 0.05) are shown. For correlation analyses (C–F), the threshold for statistically significant correlation was set at p < 0.001.

Figure 5. Kaplan-Meier Survival Curves for Total Tau (t-Tau) and Neurofilament Light (NfL) in Frontotemporal Lobar Degeneration Syndromes (FTLD-S)

Kaplan-Meier survival curves in the FTLD-S group for t-tau (A) and NfL (B). High t-tau and high NfL represent levels superior to 2.2 ng/mL and 42 ng/mL, respectively (median split).