An acetylated Tau-174 CSF biomarker discriminates between TDP-43 and tau pathology in patients with frontotemporal lobar degeneration

Madison I J Honey; Yanaika S Hok-A-Hin; Elisabeth H Thijssen; Katheryn A Q Cousins; Babette van der Weijden; Lisanne in ‘t Veld; Erik Stoops; Sterre C M de Boer; Flora H Duits; Julie F H de Houwer; Tjado H J Morrema; Daniel Alcolea; Ignacio Illán-Gala; Juan Fortea; Alberto Lleó; Adam L Boxer; David J Irwin; Edward B Lee; Leslie Shaw; Corey T McMillan; David A Wolk; Jeroen J M Hoozemans; Wiesje M van der Flier; Emma van der Ende; Harro Seelaar; Inge M W Verberk; Li Gan; Yolande Pijnenburg; Charlotte E Teunissen

doi:10.1038/s41591-026-04341-6

. 2026 Apr 15;32(6):2083–2095. doi: 10.1038/s41591-026-04341-6

An acetylated Tau-174 CSF biomarker discriminates between TDP-43 and tau pathology in patients with frontotemporal lobar degeneration

Madison I J Honey ¹, Yanaika S Hok-A-Hin ¹, Elisabeth H Thijssen ¹, Katheryn A Q Cousins ², Babette van der Weijden ¹, Lisanne in ‘t Veld ¹, Erik Stoops ³, Sterre C M de Boer ^4,⁵, Flora H Duits ⁴, Julie F H de Houwer ⁶, Tjado H J Morrema ⁷, Daniel Alcolea ^8,⁹, Ignacio Illán-Gala ^8,⁹, Juan Fortea ^8,⁹, Alberto Lleó ^8,⁹, Adam L Boxer ¹⁰, David J Irwin ², Edward B Lee ¹¹, Leslie Shaw ¹¹, Corey T McMillan ², David A Wolk ², Jeroen J M Hoozemans ⁷, Wiesje M van der Flier ^4,¹³, Emma van der Ende ¹, Harro Seelaar ⁶, Inge M W Verberk ¹, Li Gan ¹², Yolande Pijnenburg ⁴, Charlotte E Teunissen ^1,^✉

PMCID: PMC13279261 PMID: 41986736

Abstract

Biomarkers to determine underlying frontotemporal lobar degeneration (FTLD) tau or TAR DNA-binding protein (TDP) pathology during life are needed to advance clinical trials targeting specific FTD pathologies. For this purpose, we developed a new ultrasensitive immunoassay to quantify acetylated tau at lysine 174 (AcTau174) in cerebrospinal fluid (CSF). In a sporadic cohort (n = 513), AcTau174 concentrations were higher in all dementia groups (FTLD-TDP, FTLD-Tau, Alzheimer’s disease (AD), mild cognitive impairment (MCI)-AD and dementia with Lewy bodies (DLB)) compared to controls. The largest increase was observed in the FTLD-TDP group, particularly patients with semantic variant primary progressive aphasia (svPPA) and GRN mutation carriers. Notably, AcTau174 discriminated FTLD-TDP from FTLD-Tau (area under the curve (AUC) = 0.83, 95% confidence interval (CI) = 0.75–0.91) and FTLD-TDP from controls (AUC = 0.95, 95% CI = 0.92–0.99) with high accuracy. This was replicated in independent, sporadic and genetic validation cohorts (164 patients and 24 controls), albeit with somewhat lower accuracy (FTLD-TDP versus FTLD-Tau; AUC range = 0.75–0.79) and wider CIs. Within the FTLD-TDP, AD and MCI-AD groups, higher AcTau174 concentrations were associated with a faster cognitive decline over time. In summary, CSF AcTau174 has great potential to discriminate FTLD-TDP from FTLD-Tau as a biomarker reflecting FTLD-TDP disease severity and progression.

Subject terms: Dementia, Diagnostic markers, Prognostic markers, Neurodegenerative diseases, Dementia

A new cerebrospinal fluid biomarker, AcTau174, was elevated in frontotemporal lobar degeneration with TAR DNA-binding protein (FTLD-TDP), distinguishing this pathology from FTLD-Tau, and was associated with disease severity and progression in FTLD-TDP.

Main

Frontotemporal dementia (FTD) is a neurodegenerative disorder that encompasses a group of clinically heterogeneous syndromes characterized by prominent behavioral, executive function and/or language impairments^1,2. Pathologically, FTD is characterized by frontotemporal lobar degeneration (FTLD) with either TAR DNA-binding protein 43 aggregates (FTLD-TDP), tau (FTLD-Tau) or FET (FTLD-FET) protein inclusions^3,4. Mutations in the ‘progranulin’ (GRN) and ‘chromosome 9 open reading frame 72’ (C9orf72) genes lead to FTLD-TDP, while mutations in the ‘microtubule-associated protein tau’ (MAPT) gene cause FTLD-Tau⁵. However, most FTD cases are sporadic⁶, with poor correlations between the clinical syndrome and underlying pathology, except for FTD–amyotrophic lateral sclerosis (ALS) and primary progressive aphasia semantic (svPPA; likely FTLD-TDP) and progressive supranuclear palsy (PSP), corticobasal syndrome (CBS) and nonfluent variant primary progressive aphasia (nfvPPA; likely FTLD-Tau)³. In most sporadic FTD cases, molecular pathology can only be determined postmortem². Thus, these patients are excluded from clinical trials targeting abnormal TDP-43 or tau accumulation, which leads to recruitment challenges for such trials, due to the rarity of familial FTD^7,8.

To discriminate FTLD-TDP and FTLD-Tau subtypes during life, studies have explored the use of cerebrospinal fluid (CSF) and blood biomarkers, including total or phosphorylated TDP-43 and tau, neurofilament light (NfL) or glial fibrillary acidic protein^9–18. However, discriminative performance was often modest and has shown limited replication across study cohorts^9,11,19–21. Alternative biomarkers for the differentiation of FTLD-Tau from FTLD-TDP, such as TDP and tau measurement in plasma extracellular vesicles²², measurement of hepatoma-derived growth factor-like protein 2 cryptic splice variant²³ or assaying TDP seeding activity^24,25 have not yet been replicated across laboratories to date, which may be due to complexity in assaying methods. This highlights the need for new commutable biomarkers to reliably determine underlying FTLD pathology²⁶.

With no established biomarkers to distinguish tau from TDP-43, we sought new forms of tau for this purpose. Acetylation has emerged as a key post-translational modification implicated in the normal and pathological roles of tau^27–36. In particular, tau acetylation was elevated in primary neurons harboring a pathogenic FTLD-Tau mutation (P301L)³⁰. Furthermore, in cellular and mouse models of FTD, hyperacetylation of tau at lysine (K) residue 174 (AcTau174) promoted tau accumulation by inhibiting its degradation^30,37, and led to behavioral deficits in vivo, including impairments in memory and spatial learning³⁶. These findings support the potential of acetylated tau as a fluid biomarker, especially in FTLD-Tau.

We hypothesized that measuring AcTau174 could serve as a promising biomarker for FTLD-Tau. In contrast to our expectations, analyses of a pilot cohort presented in this study revealed higher concentrations of CSF AcTau174 in FTLD-TDP compared to FTLD-Tau. This study, therefore, aims to establish the potential of CSF AcTau174 as a biomarker to distinguish FTLD-TDP from FTLD-Tau, using a newly developed ultrasensitive immunoassay. In addition, using a large CSF cohort encompassing diverse neurodegenerative diseases, we thoroughly evaluated the relation of AcTau174 with various relevant clinical features in dementias. Finally, to understand its link to pathology, we characterized AcTau174 expression in postmortem brain tissue from patients with dementia.

Results

AcTau174 prototype assay and analytical validation

To test the discriminative ability of AcTau174, we measured a pilot cohort of CSF samples (Extended Data Table 1) using a prototype version of the AcTau174 assay. Contrary to our hypothesis, we observed that AcTau174 concentrations in the FTLD-TDP group were significantly higher than in the FTLD-Tau and control groups (Extended Data Fig. 1).

Extended Data Table 1.

Baseline demographics of the pilot cohort measured with the prototype AcTau174 assay

Open in a new tab

Age is shown as mean (s.d.); AcTau174 concentrations are shown as median (interquartile range). Group differences were assessed using two-sided chi-squared tests for categorical variables and Kruskal–Wallis tests for continuous variables, followed by Dunn’s post hoc tests with Bonferroni correction for multiple comparisons. Continuous biomarker concentrations were analyzed using two-sided ANCOVA adjusted for age and sex, with Bonferroni-adjusted pairwise comparisons. Abbreviations: AcTau174, acetylated Tau 174; AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; MCI, mild cognitive impairment; pTau181, phosphorylated Tau 181.

^aIndicates a statistically significant difference between groups (p < 0.05) with controls in post hoc pairwise comparisons.

^bp < 0.05 versus DLB.

^cp < 0.05 versus AD.

^dp < 0.05 versus FTLD-TDP.

^ep < 0.05 versus FTLD-Tau.

Extended Data Fig. 1 — CSF AcTau174 per clinical diagnosis (controls, n = 14; DLB, n = 26; AD, n = 37; FTLD-TDP, n = 13; FTLD-Tau, n = 10) measured using the prototype assay. Box plots display median Actau174 concentrations (centerline), with the lower and upper bounds of the box representing the first (Q1) and third (Q3) quartiles, respectively. Whiskers extend to 1.5× the interquartile range (IQR). Individual data points are shown as dots. Differences in AcTau174 concentrations between clinical groups were evaluated by a two-sided ANCOVA adjusted for age and sex, using in transformed AcTau. This was followed by post hoc pairwise comparison, which was corrected for family-wise error rate using the Bonferroni method. Abbreviations: AcTau174, Acetylated Tau 174; AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43.

Based on these results, we further optimized the assay, and technical validation of the fully developed assay demonstrated good precision (repeatability coefficient of variation (CV) of 9% and intermediate precision CV of 15%), dilution linearity (91%; Extended Data Fig. 2a,b), parallelism (95%; Extended Data Fig. 2c) and sensitivity (LLOD; 21.4 fg ml⁻¹). The anti-AcTau174 antibody demonstrated high specificity for acetylation at K174 compared with K163 or K180, and showed no binding in the absence of acetylation. Binding to K174 was not impaired by phosphorylation at neighboring threonine residue 175 (Extended Data Fig. 2d).

Extended Data Fig. 2 — a, Three CSF samples were spiked with 2,000 pg ml⁻¹ of calibrator protein and serially diluted up to a dilution factor of 2048. The mean %linearity of the assay ranged from 83% to 102%. b, No hook effect was observed for the assay based on average enzyme per bead (AEB) signals of three CSF samples measured in a serial dilution. c, The AcTau174 signal in CSF samples (log-transformed AEB) following a 1.5-fold serial dilution (reciprocal relative dilution, log-transformed) was parallel to the signal obtained from the standard curve. d, Specificity testing of the anti-AcTau174 antibody (1B8C) using synthetic biotinylated tau peptides (amino acid range 158–185) demonstrated specific binding to lysine residue 174.

We additionally developed an assay for detection in plasma; however, no difference was observed among clinical groups, nor was there a correlation between CSF and plasma AcTau174 concentrations (Supplementary Fig. 1).

Participant characteristics

The primary clinical cohort comprised 513 CSF samples from the Amsterdam Dementia Cohort (ADC; Table 1). The FTLD-TDP group (n = 52) included autopsy-confirmed cases (n = 8), individuals with a pathogenic C9orf72 (n = 7) or GRN (n = 1) mutation, and patients with svPPA (n=32) and FTD–ALS (n=4). The FTLD-Tau group (n = 79) included autopsy-confirmed cases (n = 6), individuals with a pathogenic MAPT (n = 1) mutation, along with PSP (n = 46), CBS (n = 17) and nfvPPA (n = 9) patients. The control group (n = 109) included individuals with a primary psychiatric disorder (PPD) or with subjective cognitive decline (SCD), who tested negative for Alzheimer’s disease (AD) core 1 biomarkers.

Table 1.

Baseline demographics and clinical characteristics of the extended ADC clinical cohort and the UPenn, SPIN and FTD-RisC validation cohorts

ADC (n = 513)
–	Control (n = 109)	DLB (n = 31)	MCI-AD (n = 97)	AD (n = 145)	FTLD-TDP (n = 52)	FTLD-Tau (n = 79)
Sex, M/F (%)	52/57 (48%/52%)^a	30/1 (97%/3%)^b^,c,d,e,^f	48/49 (50%/50%)^a	73/72 (50%/50%)^a	30/22 (58%/42%)^a	43/36 (54%/46%)^a
Age, year	63.9 (4)^a^,^c^,^d	69.5 (7)^b^,e,f	67.2 (5)^b^,e	67.4 (5)^b^,e	63.7 (7)^a^,c,d	65.8 (5)^a
Education, year	11.0 (3)	11.4 (3)	11.6 (3)	11.2 (3)	10.5 (2)	10.7 (3)
MMSE	27 (2)^a^,d,e,f	23 (3)^b,^c	26 (2)^a^,d,e,f	21 (5)^b^,c,f	22 (6)^b^,c	24 (4)^b^,c,d
n	108	31	96	144	50	76
FAB	16 (2)^a^,d,e,f	11 (3)^b^,c,e	16 (2)^a^,d,e,f	13 (4)^b^,c	14 (3)^a,b^,c	13 (4)^b^,c
n	101	23	84	118	35	64
FTLD-CDR SOB	–	–	–	–	6 (3)^f	5 (4)^e
n	–	–	–	–	51	77
CSF pTau181/Aβ42, Pos/Neg (%)	0/109 (0%/100%)^a^,c,d,e,f	22/9 (71%/29%)^b^,c,d,e,f	97/0 (100%/0%)^a,b^,e,f	145/0 (100%/0%)^a,b,e,f	17/35 (33%/67%)^a,b,^c,d	16/63 (20%/80%)^a,b,^c,d
n	109	31	97	145	52	79
CSF pTau181, (pg ml⁻¹)	14.1 (11.1–16.3)^a^,c,d,e	18.1 (13.7–23.4)^b^,c,d	27.9 (20.8–40.5)^a,b,^d,e,f	35.5 (24.7–47.6)^a,b,^c,e,f	15.5 (12.8–23.1)^b^,c,d	14.1 (10.8–20.3)^c,d
n	109	31	97	145	52	79
CSF Tau (pg ml⁻¹)	154 (133–178)^a^,c,d,e	189 (168–218)^b^,c,d	284 (208–379)^a,b^,d,e,f	331 (264–461)^a,b^,c,e,f	222 (175–295)^b^,c,d,f	164 (138–223)^c,d,e
n	109	31	97	145	52	79
CSF Aβ42 (pg ml⁻¹)	1,670 (1,365–1,881)^a^,c,d,f	1,232 (852–1,470)^b^,c,d	722 (607–920)^a,b^,e,f	707 (500–820)^a,b^,e,f	1,516 (1,089–1,738)^c,d	1,383 (1,117–1,719)^b^,c,d
n	109	31	97	145	52	79
CSF NfL (pg ml⁻¹)	–	–	–	–	2,911 (2,326–5,378)	2,603 (2,187–3,875)
n	–	–	–	–	25	43
Serum NfL (pg ml⁻¹)	10.5 (9.5–11.4)^e,f	15.5 (11.7–22.3)^e	12.4 (12–12.8)	19.1 (15.1–25.2)	33.1 (13.8–62.8)^a,b	21.8 (19.5–32.3)^b
n	9	29	2	38	15	35
CSF AcTau174 (fg ml⁻¹)	83.7 (51.3–103)^a^,c,d,e,f	114 (76.6–171)^b^,d,e	182 (133–280)^b^,d,e	296 (212–400)^a,b^,c,e,f	479 (277–712)^a,b^,c,d,f	138 (94.8–192)^b^,d,e
n	109	31	97	145	52	79

UPenn (n = 68)
–	Control (n = 4)	–	–	–	FTLD-TDP (n = 46)	FTLD-Tau (n = 18)
Sex, M/F (%)	2/2 (50%/50%)	–	–	–	19/27 (41%/59%)	8/10 (44%/56%)
Age, year	76.0 (3.5)^e	–	–	–	62.6 (7.4)^b	64.6 (11.1)
MMSE	30 (1)^e	–	–	–	25 (4)b	25 (5)
n	3	–	–	–	35	10
FTLD-CDR SOB	–	–	–	–	7 (5)	6 (2)
n	–	–	–	–	35	13
CSF pTau181 (pg ml⁻¹)	–	–	–	–	38.5 (25.2–45.1)	31.0 (20.4–38.0)
n	–	–	–	–	20	14
CSF Tau (pg ml⁻¹)	–	–	–	–	376 (305–548)^f	257 (206–362)^e
n	–	–	–	–	20	14
CSF NfL (pg ml⁻¹)	–	–	–	–	4,140 (1,941–7,630)	2,237 (1,451–3,259)
n	–	–	–	–	21	15
CSF AcTau174 (fg ml⁻¹)	65.0 (44.5–80.9)^e	–	–	–	273 (179–422)^b^,f	141 (48.8–202)^e
n	4	–	–	–	46	18

SPIN (n = 79)
	Control (n = 20)	–	–	AD (n = 20)	FTLD-TDP (n = 20)	FTLD-Tau (n = 19)
Sex, M/F (%)	11/9 (55%/45%)	–	–	10/10 (50%/50%)	9/11 (45%/55%)	6/13 (32%/68%)
Age, year	54.6 (11.9)^d,e,f	–	–	74 (5.1)^b^,e	63.5 (7.4)^b^,d	69.7 (8.9)^b
MMSE	30 (1)^d,e,f	–	–	19 (7)^b	23 (3)^b	25 (5)^b
n	20	–	–	11	11	14
FTLD-CDR SOB	–	–	–	NA	6 (4)	7 (5)
n	–	–	–	NA	12	17
CSF A, Pos/Neg (%)	0/20 (0%/100%)^d	–	–	20/0 (100%/0%)^b^,e,f	1/15 (6%/94%)^d	2/16 (11%/89%)^d
n	20	–	–	20	16	18
CSF T, Pos/Neg (%)	0/20 (0%/100%)^d	–	–	20/0 (100%/0%)^b^,e,f	1/16 (6%/94%)^d	2/17 (11%/89%)^d
n	20	–	–	20	17	19
CSF Aβ42 (pg ml⁻¹)	1,152 (812–1,280)^d,f	–	–	582 (479–693)^b^,e	987 (731–1,080)^d	736 (584–892)^b
n	20	–	–	20	17	18
CSF pTau181 (pg ml⁻¹)	31.5 (24.1–36.8)^d	–	–	95.7 (74–125)^b^,e,f	35.7 (28.4–39.3)^d	30.6 (24.2–34.2)^d
n	20	–	–	20	15	19
CSF Tau (pg ml⁻¹)	204 (156–286)^d	–	–	647 (530–876)^b^,e,f	289 (214–393)^d	245 (211–297)^d
n	20	–	–	20	16	19
CSF AcTau174 (fg ml⁻¹)	86.1 (71.9–134)^d,e	–	–	311 (253–474)^b^,f	253 (165–412)^b^,f	164 (86.3–231)^d,e
n	20	–	–	20	20	19

FTD-RisC (n = 41)
	C9orf72		GRN		MAPT
	Presymptomatic (n = 17)	Symptomatic (n = 10)	Presymptomatic (n = 7)	Symptomatic (n = 3)	Presymptomatic (n = 3)	Symptomatic (n = 1)
Sex, M/F (%)	6/11 (35%/65%)	6/4 (60%/40%)	3/4 (43%/57%)	0/3 (0%/100%)	2/1 (67%/33%)	1/0 (100%/0%)
Age, year	49 (13)	65 (8)	55 (10)	59 (6)	48 (3)	45 (NA)
MMSE	29 (1)	25 (7)	29 (1)	21 (7)	29 (2)	28 (NA)
n	12	8	5	3	3	1
FAB	18 (1)	12 (5)	17 (2)	10 (7)	17 (2)	16 (NA)
n	17	10	7	3	3	1
CSF AcTau174 (fg ml⁻¹)	110 (78.2–149)	263 (167–427)	63.9 (40.2–206)	847 (737–1,113)	96.3 (69.7–174)	369 (NA)
n	17	10	7	3	3	1

Open in a new tab

^aP < 0.05 versus DLB.

^bStatistically significant difference among groups (P < 0.05) with controls in post hoc pairwise comparisons.

^cP < 0.05 versus MCI-AD.

^dP < 0.05 versus AD.

^eP < 0.05 versus FTLD-TDP.

^fP < 0.05 versus FTLD-Tau.

In the ADC, the FTLD-TDP group included autopsy-confirmed cases (n = 8) and individuals with a pathogenic C9orf72 (n = 7) or GRN (n = 1) mutation. This FTLD-TDP group was further enriched for patients with svPPA (n = 32) and FTD–ALS (n = 4). The FTLD-Tau group included autopsy-confirmed cases (n = 6) and individuals with a pathogenic MAPT (n = 1) mutation. This FTLD-Tau group was further enriched for patients with PSP (n = 46), CBS (n = 17) and nfvPPA (n = 9). In UPenn, the FTLD-TDP group included individuals with a pathogenic C9orf72 (n = 19) or GRN (n = 5) mutation. This FTLD-TDP group was further enriched for patients with svPPA (n = 16), FTD–ALS (n = 4) and ALS (n = 2). Individuals in the FTLD-Tau group included those with a pathogenic MAPT (n = 3) mutation. This FTLD-Tau group was further enriched for patients with PSP (n = 15). In SPIN, the FTLD-TDP group included individuals with a pathogenic C9orf72 (n = 11) or GRN (n = 4) mutation. This FTLD-TDP group was further enriched for patients with FTD–ALS (n = 5). The FTLD-Tau group included individuals with a clinical phenotype of PSP (n = 19). Values shown as mean (s.d.), fluid biomarker values shown as median (interquartile range). Group differences were assessed using two-sided chi-squared tests for categorical variables and Kruskal–Wallis tests for continuous variables, followed by Dunn’s post hoc tests with Bonferroni correction for multiple comparisons. Continuous biomarker concentrations were analyzed using two-sided ANCOVA adjusted for age and sex, with Bonferroni-adjusted pairwise comparisons. In FTD-RisC, due to small group sizes, comparisons between the FTLD-TDP and FTLD-Tau groups could not be performed.

FAB, Frontal Assessment Battery; FTLD-CDR SOB, FTLD-modified clinical dementia rating sum-of-boxes; MTA, medial temporal lobe atrophy.

Validation cohorts included both genetic and sporadic patients with FTD, as described in the methods; the University of Pennsylvania Frontotemporal Degeneration Center (UPenn) cohort (controls (n = 4), FTLD-TDP (n = 46) and FTLD-Tau (n = 18); Table 1) and the Sant Pau Initiative on Neurodegeneration SPIN cohort (controls (n = 20), AD (n = 20), FTLD-TDP (n = 20) and FTLD-Tau (n = 19); Table 1). We also evaluated the FTD risk cohort (FTD-RisC) genetic cohort that included 27 presymptomatic individuals and 14 symptomatic individuals (Table 1).

When comparing FTLD-Tau and FTLD-TDP groups within each cohort, neuropsychological test scores were similar, except for the ADC, where the FTLD-TDP group had a higher score in the FTLD-CDR behavior/comportment/personality domain than the FTLD-Tau group (P < 0.01). CSF biomarkers (phosphorylated Tau 181 (pTau181), total tau (tTau), amyloid β 42 (Aβ42) and NfL) were available for most participants in the ADC, UPenn and SPIN cohorts. Within the ADC and uPenn cohorts, CSF tTau concentrations were significantly higher in the FTLD-TDP compared to the FTLD-Tau groups (ADC, P < 0.001; uPenn, P < 0.05); otherwise, biomarker concentrations were similar among the groups. Comparisons between FTLD-TDP and FTLD-Tau groups were not possible in the FTD-RisC due to the small number of patients with FTLD-Tau.

CSF AcTau174 is elevated in FTLD-TDP compared to all other clinical groups

In the ADC, AcTau174 concentrations were higher in all clinical groups compared to controls, with a stepwise increase from controls to DLB (1.4-fold), to MCI-AD (2.2-fold) to AD dementia (3.5-fold; Fig. 1a and Table 1). The greatest increase compared to controls was observed in the FTLD-TDP group (5.7-fold), wherein AcTau174 concentrations were significantly higher in the FTLD-TDP group compared to the FTLD-Tau group (3.5-fold). Within the FTLD-TDP group, individuals with svPPA or a GRN mutation visually exhibited the highest AcTau174 concentrations (Fig. 1b). This pattern was confirmed in an FTLD-TDP subgroup analysis, in which the elevations for svPPA (P = 0.02) and GRN carriers (P = 0.03) were statistically significant compared to C9orf72 carriers. Within the FTLD-Tau group, individuals with autopsy-confirmed PSP and Pick’s disease had consistently low AcTau174 concentrations (Fig. 1b). In the total cohort, AcTau174 concentrations were not associated with age (P = 0.81), but were slightly lower in males (P = 0.002; Table 2). After adjusting for measures of disease severity (Mini-Mental State Examination (MMSE) in the total cohort, or FTLD-CDR sum-of-boxes (SOB) in patients with FTD) in addition to age and sex, the difference between the FTLD-TDP and FTLD-Tau groups remained significant (P < 0.0001; Table 2). The FTLD-TDP versus FTLD-Tau difference remained in sensitivity analyses restricted to the subset of individuals with genetically or autopsy-confirmed FTLD pathology, in which individuals with TDP pathology type C or unclassified TDP subtypes showed the highest AcTau174 concentrations (Extended Data Fig. 3). Similar results were obtained when analyses were restricted to FTLD cases with a negative CSF AD biomarker profile (that is, no AD copathology; Extended Data Fig. 4), and AcTau174 concentrations in FTLD-TDP were comparable between individuals with a positive (median = 511 fg ml⁻¹) or negative (446 fg ml⁻¹) AD biomarker profile (P = 0.55). In the ADC, AcTau174/tTau ratio were comparable between the AD and FTLD-Tau groups, but were markedly increased in the FTLD-TDP group (Fig. 1h).

Fig. 1 — a–f, AcTau174 concentrations for all clinical groups of the ADC (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79; a), UPenn (controls, n = 4; FTLD-TDP, n = 46; FTLD-Tau, n = 18; c) and SPIN (controls, n = 20; AD, n = 20; FTLD-TDP, n = 20; FTLD-Tau, n = 19; e) cohorts. Differences in AcTau174 concentrations between clinical groups were evaluated by two-sided ANCOVA adjusted for age and sex, using natural logarithm-transformed AcTau174. This was followed by post hoc pairwise comparison, which was corrected for family-wise error rate using the Bonferroni method. AcTau174 concentrations in only FTLD-TDP and FTLD-Tau groups of the ADC (FTLD-TDP, n = 52; FTLD-Tau, n = 79; b), UPenn (FTLD-TDP, n = 46; FTLD-Tau, n = 18; d) and SPIN (FTLD-TDP, n = 20; FTLD-Tau, n = 19; f) cohorts. Data points are colored per clinical syndrome, pathology confirmation or genetic mutation. g, AcTau174 is plotted per clinical status and genetic mutation in the FTD-RisC cohort, alongside the control group of the ADC (controls, n = 109; *C9orf72* presymptomatic, n = 17; *C9orf72* symptomatic, n = 10; *GRN* presymptomatic, n = 7; *GRN* symptomatic, n = 3; *MAPT* presymptomatic, n = 3; *MAPT* symptomatic, n = 1). Significance was not tested for the *MAPT* group due to the small group numbers. h, The AcTau174/tTau ratio for all clinical groups of the ADC (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79). Differences in the AcTau174/tTau ratio between clinical groups were evaluated by two-sided ANCOVA adjusted for age and sex, using natural logarithm-transformed AcTau174. This was followed by post hoc pairwise comparison, which was corrected for family-wise error rate using the Bonferroni method. The centerline of the boxplots displays median AcTau174 concentrations (a–g), or the median AcTau174/tTau ratio (h), with the lower and upper bounds of the box representing the first (Q1) and third (Q3) quartiles, respectively. Whiskers extend to 1.5× IQR. Individual data points are shown as dots. For visualization, one data point of an individual with svPPA and an AcTau174 concentration of 2,613 fg ml⁻¹ is excluded from plots in c and d. IQR, interquartile range; tTau, total tau; ANCOVA, analysis of covariance.

Table 2.

Linear regression models of CSF AcTau174 by diagnostic group in the ADC

–	β	s.e.	P value
Model 1—diagnostic group adjusted for age and sex
Intercept	6.245	0.360	3.63 × 10⁻⁵³
Control versus FTLD-TDP	−1.765	0.105	7.40 × 10⁻⁵¹
FTLD-Tau versus FTLD-TDP	−1.075	0.111	2.54 × 10⁻²⁰
DLB versus FTLD-TDP	−1.207	0.146	1.19 × 10⁻¹⁵
MCI-AD versus FTLD-TDP	−0.829	0.109	1.14 × 10⁻¹³
AD versus FTLD-TDP	−0.347	0.102	0.0008
Age	−0.001	0.005	0.814
Sex (male versus female)	−0.179	0.056	0.002
Model 2—diagnostic group adjusted for age, sex and MMSE
Intercept	6.814	0.386	2.07 × 10⁻⁵⁴
Control versus FTLD-TDP	−1.633	0.109	5.83 × 10⁻⁴²
FTLD-Tau versus FTLD-TDP	−1.074	0.112	4.68 × 10⁻²⁰
DLB versus FTLD-TDP	−1.200	0.144	8.70 × 10⁻¹⁶
MCI-AD versus FTLD-TDP	−0.718	0.112	2.93 × 10⁻¹⁰
AD versus FTLD-TDP	−0.385	0.103	0.0002
Age	−0.001	0.005	0.885
Sex (male versus female)	−0.160	0.056	0.004
MMSE	−0.028	0.007	8.11 × 10⁻⁵
Model 3—diagnostic group adjusted for age, sex and FTLD-CDR SOB (patients with FTD only)
Intercept	6.850	0.734	5.66 × 10⁻¹⁶
FTLD-Tau versus FTLD-TDP	−1.033	0.137	8.49 × 10⁻¹²
Age	−0.014	0.011	0.199
Sex (male versus female)	−0.175	0.134	0.195
FTLD-CDR SOB	0.040	0.019	0.042

Open in a new tab

Associations between log-transformed CSF AcTau174 and diagnostic group were evaluated using linear regression, with the FTLD-TDP group as the reference category. Models included age and sex (model 1), age, sex and MMSE (model 2) or age, sex and FTLD-CDR SOB (model 3) as covariates. Model 3 was restricted to patients with FTD only, as FTLD-CDR SOB is a disease-specific measure of global severity. All P values are from two-sided tests.

Extended Data Fig. 3 — CSF AcTau174 measurements in autopsy- or genetically-confirmed FTLD-TDP (n = 16) and FTLD-Tau (n = 7) cases of the ADC. Differences in natural logarithm-transformed AcTau174 concentrations between groups were evaluated by a two-sided ANCOVA adjusted for age and sex. Box plots display median AcTau174 concentrations (centerline), with the lower and upper bounds of the box representing the first (Q1) and third (Q3) quartiles, respectively. Whiskers extend to 1.5× the interquartile range (IQR). Individual data points are shown as dots. Abbreviations: AcTau174, Acetylated Tau 174; AD, Alzheimer’s disease; *C9orf72*, chromosome 9 open reading frame 72; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; *GRN*, progranulin; *MAPT*, Microtubule-associated protein tau; PiD, Pick’s disease; PSP, progressive supranuclear palsy.

Extended Data Fig. 4 — AcTau174 concentrations for all clinical groups of the ADC (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 35; FTLD-Tau, n = 63), excluding FTLD-TDP or FTLD-Tau cases with a positive CSF AD biomarker profile (pTau181/Aβ42 ratio ≥0.02 (Elecsys) or ≥0.054 (Innotest)). Differences in natural logarithm-transformed AcTau174 concentrations between clinical groups were evaluated by a two-sided ANCOVA adjusted for age and sex. This was followed by post hoc pairwise comparison, which was corrected for family-wise error rate using the Bonferroni method. Box plots display median AcTau174 concentrations (centerline), with the lower and upper bounds of the box representing the first (Q1) and third (Q3) quartiles, respectively. Whiskers extend to 1.5× the interquartile range (IQR). Abbreviations: AcTau174, Acetylated Tau 174; AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; MCI, mild cognitive impairment.

Consistent with the ADC results, in the UPenn and SPIN validation cohorts, we observed that CSF AcTau174 concentrations were higher in the FTLD-TDP groups compared to the FTLD-Tau groups (Fig. 1c–f). In the FTD-RisC, we found that AcTau174 concentrations were higher in symptomatic FTLD-TDP individuals than in presymptomatic individuals. Notably, presymptomatic individuals had AcTau174 concentrations in a range similar to that of controls from the ADC (Fig. 1g).

CSF AcTau174 robustly discriminates FTLD-TDP from FTLD-Tau and controls

In the ADC, CSF AcTau174 effectively distinguished FTLD-TDP from FTLD-Tau with a high area under the receiver operating characteristic curve (area under the curve (AUC)) = 0.83, 95% confidence interval (CI) = 0.75–0.91; Fig. 2). Group differences in CSF pTau181, tTau, NfL and serum NfL, as well as the CSF ratios pTau181/tTau and AcTau174/pTau181 are shown in Extended Data Fig. 5. AcTau174 was significantly better than CSF pTau181 (AUC = 0.59, 95% CI = 0.49–0.69), tTau (AUC = 0.70, 95% CI = 0.61–0.79) and the CSF pTau181/tTau ratio (AUC = 0.56, 95% CI = 0.46–0.66) in discriminating FTLD-TDP from FTLD-Tau. The CSF ratios AcTau174/pTau181 and AcTau174/tTau did not improve discriminative accuracy compared to AcTau174 alone (AcTau174/pTau181 AUC = 0.80, 95% CI = 0.72–0.88; AcTau174/tTau AUC = 0.81, 95% CI = 0.74–0.89). In a subset of individuals with CSF NfL (n = 68) or serum NfL (n = 50) measurements, AcTau174 additionally outperformed NfL for the discrimination of FTLD-TDP from FTLD-Tau (CSF NfL AUC = 0.57, 95% CI = 0.42–0.71; serum NfL AUC = 0.59, 95% CI = 0.37–0.80). AcTau174 demonstrated high accuracy for distinguishing FTLD-TDP from controls (AUC = 0.95, 95% CI = 0.92–0.99), again outperforming CSF pTau181, tTau and pTau181/tTau. AcTau174 distinguished FTLD-Tau from controls with an AUC of 0.79 (95% CI = 0.72–0.86), which was significantly better than CSF pTau181, tTau and pTau181/tTau. Our findings were replicated in the validation cohorts; AcTau174 distinguished FTLD-TDP from FTLD-Tau with good, albeit somewhat lower, accuracies with wider confidence intervals, with an AUC of 0.79 in UPenn (95% CI = 0.68–0.91) and 0.75 in SPIN (95% CI = 0.59–0.90; Fig. 2). In UPenn, AcTau174 showed a comparable AUC to the CSF pTau181/tTau ratio, whereas in SPIN, AcTau174 maintained the highest AUC.

Fig. 2 — a–c, ROC analysis was used to calculate the AUC (95% CI) for the discrimination among FTLD-TDP, FTLD-Tau and controls using CSF biomarkers AcTau174, tTau, pTau181 and NfL, or serum NfL in the ADC (a), SPIN (b) and UPenn (c) cohorts. ROC-AUCs are not presented for UPenn FTLD-TDP versus controls, and FTLD-Tau versus controls, due to the small size of the control group (n = 4). pTau181, phosphorylated Tau 181; ROC, receiver operating curve.

Extended Data Fig. 5 — a–e, Concentrations of CSF pTau181 (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79; a), CSF total tau (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79; b), CSF pTau181/total tau (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79; c), CSF AcTau174/pTau181 (controls, n = 109; DLB, n = 31; MCI-AD, n = 97; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79; d) and serum NfL (controls, n = 9; DLB, n = 29; MCI-AD, n = 2; AD, n = 38; FTLD-TDP, n = 15; FTLD-Tau, n = 35; e) for all clinical groups of the ADC where available. Differences in biomarker concentrations and ratios between clinical groups were evaluated by two-sided ANCOVA adjusted for age and sex, using natural logarithm-transformed biomarker concentrations or ratios. This was followed by post hoc pairwise comparison, which was corrected for family-wise error rate using the Bonferroni method. Box plots display median biomarker concentrations (centerline), with the lower and upper bounds of the box representing the first (Q1) and third (Q3) quartiles, respectively. Whiskers extend to 1.5× the interquartile range (IQR). Abbreviations: AcTau174, Acetylated Tau 174; AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; MCI, mild cognitive impairment; NfL, neurofilament light; pTau181, phosphorylated Tau 181.

For discriminating FTD from AD, AcTau174 performed differently across pathological subtypes. Discrimination between FTLD-TDP and AD was modest for AcTau174 alone (AUC = 0.66, 95% CI = 0.56–0.75; Extended Data Fig. 6), whereas biomarker ratios incorporating AcTau174 improved diagnostic performance. Both the AcTau174/pTau181 ratio (AUC = 0.88, 95% CI = 0.81–0.95) and the AcTau174/tTau ratio (AUC = 0.82, 95% CI = 0.74–0.91) outperformed AcTau174 alone, and showed comparable accuracy to CSF pTau181 alone (AUC = 0.85, 95% CI = 0.79–0.92). In contrast, for FTLD-Tau versus AD, AcTau174 alone showed good discrimination (AUC = 0.80, 95% CI = 0.73–0.87), whereas AcTau174-based ratios did not improve accuracy. For this comparison, CSF pTau181 alone demonstrated the highest accuracy (AUC = 0.92, 95% CI = 0.88–0.95; Extended Data Fig. 6).

Extended Data Fig. 6 — Forest plot displaying the area under the ROC curve (AUC) and 95% confidence intervals for the discrimination of different clinical groups using CSF biomarkers (AcTau174, total tau, pTau181 and CSF biomarker ratios; AD, n = 145; FTLD-TDP, n = 52; FTLD-Tau, n = 79), or serum NfL (AD, n = 38; FTLD-TDP, n = 15; FTLD-Tau, n = 35). Points represent the estimated AUC and horizontal error bars indicate the 95% confidence intervals. FTLD-TDP vs. AD two-sided DeLong’s test: AcTau174 vs. pTau181, Z = −2.93, p = 0.003, 95% CI: −0.33–0.06; AcTau174 vs. total tau, Z = −1.57, p = 0.12, 95% CI: −0.27–0.03; AcTau174 vs. pTau181/Total Tau, Z = −3.05, p = 0.002, 95% CI: −0.30–0.06; AcTau174 vs. AcTau174/pTau181, Z = −5.03, p = 4.95 × 10⁻⁷, 95% CI: −0.31–0.14; AcTau174 vs. AcTau174/total tau, Z = −5.69, p = 1.31 × 10⁻⁸, 95% CI: −0.23–0.11; pTau181 vs. AcTau174/pTau181, Z = −0.74, p = 0.46, 95% CI: −0.10–0.04 FTLD-Tau vs. AD two-sided DeLong’s test: AcTau174 vs. pTau181, Z = −3.42, p = 0.00063, 95% CI: −0.18–0.05; AcTau174 vs. total tau, Z = −3.35, p = 0.00080, 95% CI: −0.14–0.04. Abbreviations: AcTau174, acetylated Tau 174; AD, Alzheimer’s disease; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; NfL, neurofilament light; pTau181, phosphorylated Tau 181.

CSF AcTau174 reflects disease severity and neurodegeneration across cohorts

To further understand the clinical relevance of elevated CSF AcTau174 concentrations, we examined its relationship with clinical and other biomarker measures within disease (sub)groups.

Within the FTLD-TDP group of the ADC, CSF AcTau174 correlated strongly with CSF tTau (ρ = 0.64, P < 0.0001), and pTau181 (ρ = 0.49, P = 0.0002), but not Aβ42 (ρ = 0.25, P = 0.08; Supplementary Fig. 2). In the subset of individuals with CSF NfL (n = 25) or serum NfL (n = 15) measurements available, AcTau174 was strongly correlated with NfL in both matrices (CSF Nfl: ρ = 0.73, P < 0.0001; serum NfL: ρ = 0.90, P < 0.0001). In addition, AcTau174 was significantly correlated with MMSE (ρ = −0.29, P = 0.039; Supplementary Fig. 3) in the entire FTLD-TDP group. Due to heterogeneity within the FTLD-TDP group of the ADC, we performed subgroup analyses in patients with svPPA (n = 32), the largest and thus best-powered FTLD-TDP subgroup. Here, AcTau174 showed a stronger negative correlation with MMSE (ρ = −0.41, P = 0.026; Supplementary Fig. 2) as compared to the correlation between AcTau174 and MMSE in the total FTLD-TDP group. Similar to the FTLD-TDP group, in the FTLD-Tau group of the ADC, CSF AcTau174 showed significant but weaker correlations with CSF tTau (ρ = 0.61, P < 0.0001), pTau181 (ρ = 0.37, P = 0.001), CSF NfL (ρ = 0.50, P = 0.001) and serum NfL (ρ = 0.35, P = 0.04). In this group, AcTau174 additionally correlated with MMSE score (ρ = −0.23, P = 0.049; Supplementary Fig. 2).

In the UPenn FTLD-TDP group, results were comparable to the ADC; CSF AcTau174 showed significant correlations with tTau (ρ = 0.61, P = 0.004), pTau181 (ρ = 0.51, P = 0.023) and NfL (ρ = 0.72, P < 0.0001; Supplementary Table 1), but no cross-sectional associations were found between AcTau174 and MMSE or FTLD-CDR scores. In the UPenn FTLD-Tau group, AcTau174 did not correlate with CSF biomarkers or neuropsychological test scores.

In the SPIN FTLD-TDP group, we found no associations between AcTau174 and MMSE, FTLD-CDR scores or CSF biomarkers, likely due to a smaller group size (Supplementary Table 2). In the SPIN FTLD-Tau group, higher CSF AcTau174 concentrations correlated with a worse score for the FTLD-CDR behavior/comportment/personality domain (ρ = 0.52, P = 0.031), higher CSF pTau181 (ρ = 0.63, P = 0.004) and tTau (ρ = 0.48, P = 0.039; Supplementary Table 2).

Finally, within the FTLD-TDP presymptomatic group (GRN and C9orf72 carriers) of the FTD-RisC, we found a significant cross-sectional association between AcTau174 and MMSE (ρ = −0.63, P = 0.006; Supplementary Fig. 4). Within the FTLD-TDP symptomatic group, no significant associations were found between AcTau174 and cognitive test scores, possibly due to the small group size; however, we observed a trend with the FTLD-CDR SOB (ρ = 0.58, P = 0.088) and FTLD-CDR memory domain (ρ = 0.63, P = 0.051; Supplementary Table 3). We could not test associations in the FTLD-Tau presymptomatic and symptomatic groups due to the small group sizes.

Baseline CSF AcTau174 predicts future cognitive and functional decline

We then investigated whether baseline CSF AcTau174 concentrations had prognostic value across disease groups. In the ADC, higher baseline CSF AcTau174 concentrations were associated with a faster rate of cognitive decline, as measured with longitudinal MMSE, within the FTLD-TDP group (β = −1.64; 95% CI = −2.61, −0.87; P = 0.0014; Fig. 3a). Of note, MMSE appeared almost stable over 6 years in the group with AcTau174 concentrations below the median. These effects were not observed for CSF pTau181 or tTau (Extended Data Table 2). Higher baseline CSF AcTau174 was also associated with a more rapid MMSE decline within the AD group (β = −1.66; 95% CI = −2.44, −0.90; P < 0.0001) and the MCI-AD group (β = −0.67; 95% CI = −1.09, −0.27; P = 0.002) but not in the FTLD-Tau group (Fig. 3 and Extended Data Table 2). Higher CSF AcTau174 concentrations were not significantly associated with mortality in the FTLD-TDP or FTLD-Tau groups; however, there was an association with risk of death in the AD (hazard ratio = 1.5; 95% CI = 1.1, 2.1; P = 0.02) and MCI-AD groups (hazard ratio = 2.2; 95% CI = 1.1, 4.2; P = 0.02; Supplementary Table 4).

Fig. 3 — a–f, The effect of baseline AcTau174 on longitudinal changes of cognitive test scores within the ADC FTLD-TDP (MMSE) (a), ADC FTLD-Tau (MMSE) (b), UPenn FTLD-TDP (MMSE) (c), UPenn FTLD-TDP (FTLD-CDR) (d), ADC MCI-AD (MMSE) (e) and ADC AD (MMSE) (f) groups. Linear mixed-effects models with random intercepts and random slopes were used to evaluate the relationship between baseline AcTau174 and changes in MMSE or FTLD-CDR over time. Models were adjusted for age, sex, years of education, follow-up time in years and an interaction between AcTau174 and time. AcTau174 concentrations were dichotomized to low or high groups based on the median value of the clinical group. Average slopes with 95% CIs adjusted for age, sex and years of education are superimposed on the graphs.

Extended Data Table 2.

The effect of baseline CSF AcTau174, pTau181 or Tau on longitudinal changes in MMSE in ADC

Open in a new tab

Linear mixed-effects models with random intercepts and random slopes were used to evaluate the relationship between baseline biomarkers (z-scored) with changes in MMSE over time. Models were adjusted for age, sex, years of education, follow-up time in years, and an interaction between biomarker and time. All statistical tests were two-sided and no adjustment for multiple comparisons was applied. Abbreviations: AcTau174, acetylated Tau 174; AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; FTLD, frontotemporal lobar degeneration; MCI, mild cognitive impairment; MMSE, mini-mental state examination; pTau181, phosphorylated Tau 181.

Findings from the UPenn cohort supported the prognostic value of CSF AcTau174 in FTLD-TDP; higher baseline AcTau174 was associated with a faster rate of cognitive decline as measured by MMSE, although this association was borderline significant (β = −1.63; 95% CI = −3.25, −0.05, P = 0.06; Fig. 3c and Extended Data Table 3). In addition, higher baseline AcTau174 was associated with a significantly faster rate of functional and behavioral decline, measured by the FTLD-CDR (β = 1.41; 95% CI = 0.48, 2.37; P = 0.009; Fig. 3d). CSF NfL, but not pTau181 or tTau, showed similar prognostic value to AcTau174 (Extended Data Table 3).

Extended Data Table 3.

The effect of baseline CSF AcTau174, pTau181, Tau or NfL on longitudinal changes in MMSE or FTLD-CDR within the UPenn FTLD-TDP group

Open in a new tab

Linear mixed-effects models with random intercepts and random slopes were used to evaluate the relationship between baseline biomarkers (z-scored) with changes in MMSE or FTLD-CDR over time within the FTLD-TDP group. Models were adjusted for age, sex, years of education, follow-up time in years, and an interaction between biomarker and time. The FTLD-Tau group was not evaluated due to too few individuals with longitudinal data. All statistical tests were two-sided and no adjustment for multiple comparisons was applied. Bold entries refer to significant findings. Abbreviations: AcTau174, acetylated Tau 174; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; MMSE, mini-mental state examination; NfL, neurofilament light; pTau181, phosphorylated Tau 181.

AcTau174 in brain tissue suggests diagnostic relevance of soluble tau species

To understand how our CSF findings relate to acetylation of tau within the brain, we performed immunohistochemical staining and quantified expression (immunoreactivity) using the same anti-AcTau174 antibody as used in our immunoassay, on frontal, temporal, parietal and hippocampal tissue from a postmortem cohort (Extended Data Table 4). This cohort included individuals with a neuropathological diagnosis of FTLD-TDP (n = 12), FTLD-Tau (n = 17), AD (n = 25), DLB (n = 9), and non-neurodegenerative controls (n = 10; AD neuropathological change scores of 1 or below)³⁸.

Extended Data Table 4.

Demographics and neuropathological characteristics of the immunohistochemistry brain tissue cohort which was stained for AcTau174

Open in a new tab

Values shown as mean (standard deviation), neuropathological scores shown as median (interquartile range). Not every individual per pathological group had every region available for acetylated tau and phosphorylated tau staining. The FTLD-TDP group was comprised of FTLD-TDP type A (n = 5), FTLD-TDP type B (n = 2), or unclassifiable FTLD-TDP subtypes (n = 5). Abbreviations: ABC A score, Aβ/amyloid plaque score; ABC B score, neurofibrillary tangle score; ABC C score, neuritic plaque score; AcTau174, acetylated Tau 174; AD, Alzheimer’s disease; DLB, dementia with Lewy bodies; FTLD-Tau, frontotemporal lobar degeneration–Tau; FTLD-TDP, frontotemporal lobar degeneration-TDP-43; pTau202/205, Phosphorylated Tau 202/205.

^aIndicates a statistically significant difference between groups (p < 0.05) with controls in post hoc pairwise comparisons.

^bp < 0.05 versus DLB.

^cp < 0.05 versus AD.

^dp < 0.05 versus FTLD-TDP.

^ep < 0.05 versus FTLD-Tau.

Across all four brain regions, AcTau174 immunoreactivity was highest in the AD group compared to the control, DLB and FTLD-TDP groups (Fig. 4a). AcTau174 immunoreactivity was higher in the FTLD-Tau group than in the FTLD-TDP group, which was an inverse observation compared to our CSF findings. These findings suggest that TDP-43 pathology may interfere with the intracellular accumulation of AcTau174. The anti-AcTau174 antibody labeled structures morphologically consistent with neurofibrillary tangles (light blue arrow), dystrophic neurites (green arrow) and senile plaques (dark blue arrow) in AD cases, as well as neuropil threads (purple arrow) and Pick bodies (red arrow) in FTLD-Tau cases, while AcTau174 labeling in neurite-like structures appeared lower in FTLD-TDP than in controls (Fig. 4b).

To investigate the pathological correlates of AcTau174, we examined its associations with pTau202/pTau205 immunoreactivity (reflecting phosphorylated tau species) and AD neuropathological changes (reflecting amyloid plaques, neurofibrillary tau tangles and neuritic plaque scores). In the whole cohort, AcTau174 immunoreactivity was moderately positively correlated with pTau202/pTau205 immunoreactivity in both frontal and hippocampal tissue (Supplementary Fig. 5). No significant associations were found between AcTau174 immunoreactivity and amyloid plaque, neurofibrillary tangle or neuritic plaque scores in any of the four regions, in the whole cohort or per disease.

We then analyzed paired antemortem CSF and postmortem brain AcTau174 immunoreactivity from three FTLD-TDP and six AD cases (average interval between CSF collection and tissue collection at death = 4.6 ± 3.2 years). Of the four brain regions, adjusted for the CSF-death interval, hippocampal AcTau174 immunoreactivity showed a significant inverse association with CSF AcTau174 concentrations (β = −0.00197, P = 0.047).

To assess whether the associations observed in intracellular staining extend to whole tissue and soluble protein, we measured AcTau174 in brain homogenates of a subset of the FTLD-TDP and FTLD-Tau cases (Supplementary Table 5). Consistent with the CSF findings, we observed higher AcTau174 concentrations in the FTLD-TDP homogenates compared to the FTLD-Tau homogenates (P = 0.137; Fig. 4c), although this difference was not significant, likely due to limited power. AcTau174 concentrations in brain homogenates were similarly inversely correlated with AcTau174 immunoreactivity (ρ = −0.67, P = 0.039; Fig. 4d). Based on these findings, we hypothesized that elevated CSF AcTau174 concentrations in FTLD-TDP reflect the soluble protein fraction, and that the inverse association between AcTau174 immunoreactivity and CSF concentrations may arise from differences in detecting soluble versus insoluble AcTau174 across matrices.

Discussion

Our study shows that CSF AcTau174 concentrations differentiate patients with FTLD-TDP from those with FTLD-Tau across clinically, genetically, and autopsy-diagnosed groups, based on a total of 298 patients with FTD. For the discrimination of FTLD-TDP from FTLD-Tau, CSF AcTau174 outperformed commonly used AD and FTD biomarker measurements, including CSF pTau181, tTau, pTau181/tTau and NfL. AcTau174 shows prognostic potential, as it predicted the rate of cognitive and functional decline over 6 years of follow-up within patients with FTLD-TDP. In brain homogenates containing soluble protein, we observed a trend of increased AcTau174 in FTLD-TDP compared to FTLD-Tau, mirroring the results in CSF. However, immunohistochemistry revealed low AcTau174 staining in FTLD-TDP brain tissue, in contrast to the strong staining in FTLD-Tau and AD. Taken together, our data suggest a new process involving elevated soluble CSF AcTau174 concentrations in FTD cases with TDP-43 pathology, which may have clinical relevance.

Currently, in a clinical FTD setting, the AD core 1 biomarkers can be used to exclude an AD diagnosis, and NfL can be used to discriminate FTD from psychiatric disorders³⁹. AcTau174 concentrations were strongly correlated with CSF pTau181, tTau, CSF and serum NfL, raising the question of its added value. Notably, AcTau174 concentrations in FTLD-TDP were not influenced by the presence of AD copathology, suggesting that it captures FTD-related changes distinct from those reflected by established biomarkers. Thus, AcTau174 could provide complementary diagnostic value in a stepwise process; first, using current markers or panels to diagnose FTD⁴⁰, and, second, using AcTau174 to discriminate pathological subtypes. This may be valuable not only in diagnostic settings but also in clinical trial contexts for biologically enriching and stratifying patient populations in FTD trials.

Reliable prognostic biomarkers are a key unmet need in FTD, given the clinical and pathological heterogeneity, variable onset and progression, and the increasing relevance of stratification in clinical trials²⁶. In FTLD-TDP, AcTau174 showed superior prognostic value for cognitive and functional decline compared with pTau181 or tTau, as observed in multiple cohorts. Within presymptomatic FTLD-TDP in the FTD-RisC, higher AcTau174 concentrations were already associated with lower MMSE scores at baseline, but AcTau174 concentrations were not increased in presymptomatic mutation carriers compared to healthy controls. Our findings, therefore, suggest prognostic utility mainly in symptomatic disease, although the presymptomatic group was small, limiting firm conclusions.

To better contextualize the disease specificity of CSF AcTau174, it is important to distinguish its behavior across the neurodegenerative diseases we studied. We showed that CSF and brain homogenate concentrations of AcTau174 were inversely related to AcTau174 immunoreactivity, possibly due to differences in solubility. Evaluation of the CSF AcTau174/tTau ratio revealed a disease-specific pattern; while AcTau174/tTau levels were comparable in AD and FTLD-Tau, they were markedly increased in FTLD-TDP. This may indicate that, in FTLD-TDP, elevated CSF AcTau174 reflects a relative enrichment of acetylation within the soluble secreted tau pool, but in AD, CSF AcTau174 may mirror increased global tau release. In contrast, although immunostainings showed prominent tau acetylation in FTLD-Tau brains, as in AD, CSF AcTau174 concentrations were comparatively lower. Based on our postmortem findings, AcTau174 may be preferentially sequestered within insoluble tau inclusions in tauopathies such as FTLD-Tau, thereby limiting its availability in the soluble and CSF compartments. We observed a small number of high CSF AcTau174 values in the clinical FTLD-Tau group, primarily individuals with CBS or PSP. Roughly 14% PSP and 33% CBS cases have TDP-43 pathology present at autopsy⁴¹; thus, we cannot rule out that high AcTau174 concentrations in these individuals may be reflective of TDP-43 copathology driving the increase in soluble protein. In AD, although AcTau174 is also incorporated into brain tau aggregates, higher CSF AcTau174 levels may reflect distinct or additional tau acetylation pathways, potentially driven by amyloid-related mechanisms that promote enhanced tau release or turnover, as previously suggested⁴². We observed low CSF AcTau174 concentrations in DLB, a primarily non-Tau neurodegenerative disease, supporting the specificity of AcTau174 elevations to TDP-43-associated pathology. Overall, these findings support a model in which CSF AcTau174 is strongly influenced by disease-specific mechanisms regulating tau post-translational modification, solubility and secretion.

Our rationale for targeting tau acetylation at lysine 174 was twofold. From a biological perspective, acetylation at K174 has been shown to impair tau degradation, thereby promoting pathological accumulation. Its pathogenic relevance has been demonstrated in vivo, where AcTau174 exacerbated tau-mediated neurodegeneration and induced cognitive and behavioral deficits^30,36,37. From a technical standpoint, CSF tau predominantly consists of truncated species spanning the N-terminal to midregion⁴³, including the proline-rich domain in which K174 resides. This made AcTau174 more amenable to reliable quantification in CSF than acetylation sites in lower-abundance C-terminal regions. We acknowledge that more than 20 other tau lysine residues have been identified⁴⁴, some of which may be more closely linked to tau aggregation and hold relevance for FTLD-Tau. In particular, acetylation at K274 has been associated with insoluble tau aggregates in AD and FTLD-Tau brains^45–47, and has been shown to impair synaptic plasticity and memory in transgenic tau mouse models²⁸. Owing to its biological relevance, we attempted to quantify CSF AcTau274; however, sufficient assay sensitivity could not be achieved (M.I.J.H. and C.E.T., unpublished data), potentially reflecting its more distal C-terminal localization. In addition, emerging evidence suggests that acetylation at K298 may be particularly relevant for 3R tauopathies, as this site has been identified as a critical determinant in directing 3R tau agreggation⁴⁸. Future studies incorporating a broader panel of acetylated tau species may therefore improve pathological specificity, particularly for FTLD-Tau subtypes.

Our results add to emerging evidence suggesting that TDP-43 pathology is related to tau abnormalities in FTLD, including its phosphorylation and altered conformation^49,50. Within the FTLD-TDP group, the highest CSF AcTau174 concentrations were observed in individuals with svPPA or GRN mutations, with smaller elevations in C9orf72 expansion carriers. The lower burden of TDP-43 species reported in C9orf72-associated disease is consistent with our hypothesis that pathological TDP-43 may influence tau protein acetylation⁵¹. Mechanistically, pathological TDP-43 could induce broad protein acetylation by modulating lysine acetyltransferases, such as p300 and CREB-binding protein, or deacetylases, potentially through impaired RNA binding and altered regulation of acetylation-related transcripts secondary to its nuclear depletion. Knockdown of TDP-43 in human neuronal cells has been demonstrated to downregulate histone deacetylase 6 (HDAC6)⁵². HDAC6 inhibition induces tau acetylation, although this inhibition alone was not sufficient to generate acetylated tau in primary cortical neurons of mice lacking mature tau pathology⁵³. Future mechanistic studies are warranted to clarify whether pathological TDP-43 directly drives tau acetylation, whether broader protein acetylation occurs and how these mechanisms diverge across different diseases. In this context, evaluation of CSF AcTau174 in ALS, a disorder predominantly associated with TDP-43 pathology⁵⁴, may further inform whether the observed changes in FTD reflect shared TDP-43-related mechanisms.

This study has a number of limitations. Although CSF AcTau174 consistently distinguished FTLD-TDP from FTLD-Tau across cohorts, its diagnostic performance was moderate in the ADC cohort and slightly more modest in the validation cohorts, with overlap in concentrations among groups. This variability underscores that, while informative at the group level, CSF AcTau174 may have limited standalone utility for individual-level subtype classification. The diagnostic potential of AcTau174 for the discrimination of FTLD-TDP from FTLD-Tau can potentially be further improved when combined with additional biomarkers to reach a satisfactory high diagnostic accuracy, which was, however, outside the scope of the current study. Another limitation of our study is the low number of genetic FTD cases we were able to include, particularly presymptomatic individuals. We selected cohorts enriched for cases with highly predictable neuropathology, based on genetic confirmation or well-defined clinical phenotypes, and while our new biomarker AcTau174 shows encouraging performance in these FTD cases, a key unmet need for a diagnostic biomarker lies in sporadic behavioral variant patients with FTD (bvFTD), where pathology is unpredictable. To further address the aforementioned limitations, validation of the diagnostic performance of CSF AcTau174 in a larger number of cases with autopsy or genetic confirmation of pathology, including individuals with bvFTD, is warranted. Finally, the pattern we observed in CSF did not translate to plasma, limiting the potential for broad, noninvasive clinical use of AcTau174. However, plasma analyses were restricted to total AcTau174 rather than brain-derived AcTau174, and future studies using assays specific for brain-derived species and in combination with ultrasensitive analytical platforms may be required to fully assess its utility in blood.

Our study provides new evidence that CSF AcTau174 concentrations tend to be higher in individuals with symptoms of FTD associated with TDP-43 pathology and may help discriminate FTLD-TDP from FTLD-Tau, with potential prognostic relevance for patients with FTLD-TDP. The availability of such biomarkers could be valuable in supporting the development of therapies targeting the specific pathologies that characterize FTLD.

Methods

Patient cohorts

ADC

This study included individuals from the ADC⁵⁵ and SCIENCe project⁵⁶ who visited the memory clinic of the Alzheimer Center Amsterdam (Vrije Universiteit Amsterdam, Amsterdam UMC location VUmc) for diagnostic purposes between 2003 and 2024. Individuals were included in the study if they had a CSF sample collected within 6 months of their diagnostic clinic visit. During their visit, individuals underwent diagnostic workup including physical, neurological and neuropsychological evaluations, brain magnetic resonance imaging (MRI), apolipoprotein E genotyping and CSF AD biomarker (pTau181, total tau and Aβ42) analysis. CSF AD biomarkers were measured using either Elecsys assays (pTau181/Aβ42 cutoff ≥ 0.02) or Innotest assays (pTau181/Aβ42 cutoff ≥ 0.054) and values were transformed to the same scale using the formulas: Elecsys Aβ42 = −365 + 1.87 × Innotest Aβ42; Elecsys pTau181 = −3.9 + 0.44 × Innotest pTau181; Elecsys tTau = 52 + 0.42 × Innotest tTau⁵⁷. Some individuals additionally received CSF and/or serum NfL analysis, measured using the Simoa NfL assay. A consensus diagnosis was established by a multidisciplinary team following diagnostic criteria for—bvFTD², svPPA, nfvPPA, logopenic variants (lvPPA)¹, FTD–ALS⁵⁸, CBS⁵⁹, PSP⁶⁰, AD dementia⁶¹, DLB⁶² and MCI⁶³. An individual was classified as having SCD when they presented to the clinic with cognitive complaints but had no objective impairment as measured on their neuropsychological and neurological evaluations, and did not meet criteria for other neurological or psychiatric diseases. Individuals determined to have a PPD for which diagnostic uncertainty remained, or when FTD was considered in the differential diagnosis, were referred to our specialized clinic for further evaluation⁶⁴. When longitudinal follow-up confirmed a psychiatric diagnosis with certainty, these cases were retained as controls (n = 8). All other individuals with a psychiatric diagnosis that were included as controls did not have a phenotype suggestive of FTD, and adhered to the Diagnostic and Statistical Manual of Mental Disorders, fifth edition⁶⁵.

Clinical disease severity was assessed by validated tests covering the major cognitive domains. Behavioral symptoms were assessed using the Frontal Assessment Battery score (range = 0–18) and global cognition was assessed using the MMSE score (range = 0–30). In patients with FTD, global cognition and function were additionally assessed with the frontotemporal lobar degeneration-modified clinical dementia rating SOB score (range = 0–24)⁶⁶. The education level of each individual was determined using the Dutch Verhage scale and also used to calculate total years of education⁶⁷.

For the FTD group, we included individuals for whom the underlying disease pathology (FTLD-TDP or FTLD-Tau) could be determined by either the presence of a pathogenic mutation (C9orf72, GRN or MAPT mutation), autopsy confirmation, or if the individual had a clinical syndrome that is strongly associated with TDP or tau neuropathology. For the FTLD-TDP group, these clinical syndromes were svPPA and FTD–ALS^68,69. For the FTLD-Tau group, these clinical syndromes were PSP, CBS and nfvPPA^68,70. Individuals with AD and MCI-AD were included in the case of a positive CSF AD biomarker profile (pTau181/Aβ42 ratio of ≥0.02 (Elecsys) or of ≥0.054 (Innotest)). Individuals were included in the control group if they had either SCD or a PPD, and displayed a negative CSF AD biomarker profile. We excluded an individual’s neuropsychological and MRI data if it was collected more than 6 months after the date of lumbar puncture.

In the ADC pilot cohort, the control group included seven individuals with SCD and seven with psychiatric disorders. These psychiatric disorders included mood disorders (n = 2) and other psychiatric disorders (n = 5)⁶⁵. The FTLD-TDP group included autopsy-confirmed cases (n = 3) and individuals with a pathogenic C9orf72 mutation (n = 1). This FTLD-TDP group was further enriched for patients with svPPA (n = 9). The FTLD-Tau group included patients with nfvPPA (n = 8), CBS (n = 1) and PSP (n = 1).

In the extended primary ADC clinical cohort, the control group included 39 individuals with SCD and 70 with psychiatric disorders. These psychiatric disorders included mood disorders (n = 28), anxiety disorders (n = 2), functional disorders (n = 2), sleep disorders (n = 2), trauma-related disorders (n = 3), autism spectrum disorder (n = 2), substance or alcohol use disorder (n = 2) and other psychiatric disorders (n = 29)⁶⁵. The FTLD-TDP group included autopsy-confirmed cases (n = 8) and individuals with a pathogenic C9orf72 (n = 7) or GRN (n = 1) mutation. This FTLD-TDP group was further enriched for patients with svPPA (n = 32) and FTD–ALS (n = 4). The FTLD-Tau group included autopsy-confirmed cases (n = 6) and individuals with a pathogenic MAPT (n = 1) mutation. This FTLD-Tau group was further enriched for patients with PSP (n = 46), CBS (n = 17) and nfvPPA (n = 9).

FTD-RisC

Participants from the FTD-RisC study were included based on carriership of a major autosomal pathogenic mutation in either the GRN, C9orf72 or MAPT gene, as described previously^71,72. Participants underwent genotyping at baseline visit and were followed annually or bi-annually for a standardized assessment including neurological and neuropsychological examination, venipuncture and MRI. Participants were defined as presymptomatic (n = 27) when established clinical diagnostic criteria for FTD, PPA or ALS were not fulfilled, and symptomatic (n = 14) when a diagnosis was made according to the established criteria^1,2,73.

UPenn cohort

From the UPenn cohort⁷⁴, individuals in the FTLD-TDP group included those with a pathogenic C9orf72 (n = 19) or GRN (n = 5) mutation. This FTLD-TDP group was further enriched for patients with svPPA (n = 16), FTD–ALS (n = 4) and ALS (n = 2). Individuals in the FTLD-Tau group included those with a pathogenic MAPT (n = 3) mutation. This FTLD-Tau group was further enriched for patients with PSP (n = 15). The control group included individuals who were cognitively normal with a negative CSF AD biomarker profile (n = 4). All individuals underwent a standardized assessment including biospecimen collection and clinical and neuropsychological assessments, and a diagnosis was established based on the applicable clinical criteria^1,2,58,60. A subset of individuals had CSF AD biomarkers measured using the Lumipulse platform and NfL measured using the Simoa NfL assay.

SPIN cohort

From the SPIN cohort⁷⁵, individuals in the FTLD-TDP group included those with a pathogenic C9orf72 (n = 11) or GRN (n = 4) mutation. This FTLD-TDP group was further enriched for patients with FTD–ALS (n = 5). The FTLD-Tau group included individuals with a clinical phenotype of PSP (n = 19). The control group included individuals who were cognitively normal with a negative CSF AD biomarker profile (n = 20). Finally, the AD dementia group included individuals with a positive CSF AD biomarker profile (n = 20). CSF AD biomarkers were measured using the Lumipulse G System⁷⁶. All individuals underwent a standardized assessment including neurological and neuropsychological examination, venipuncture, lumbar puncture and MRI. A diagnosis was established based on the applicable clinical criteria^58–61.

The Netherlands Brain Bank postmortem cohort

Postmortem paraffin-embedded and frozen tissue blocks from patients were obtained from the Netherlands Brain Bank (NBB), the Netherlands Institute for Neuroscience, Amsterdam (open access at www.brainbank.nl). Brain autopsies were performed according to international guidelines of Brain Net Europe⁷⁷ and pathological examinations were conducted by an experienced neuropathologist. Participants were included based on availability of frontal, parietal, temporal and hippocampal tissue, with a neuropathological diagnosis of FTLD-TDP (n = 12)⁷⁸, FTLD-Tau (n = 17)⁷⁸, AD (n = 25)⁷⁹, DLB (n = 9)⁸⁰ and controls (n = 10). Not all cases had background neuropathology defined; however, when available, AD neuropathologic change was evaluated with an ‘ABC’ score³⁸ and was extracted from neuropathological reports.

Ethical approval

The studies from which data and samples were used were all approved by local Institutional Ethical Review Boards (that is, the Amsterdam University Medical Centers, the Erasmus Medical Center, the University of Pennsylvania Perelman School of Medicine and the Hospital de la Santa Creu i Sant Pau). All participants provided written informed consent for the use of their medical data and biospecimens for research. The Institutional Ethical Review Board of the Amsterdam University Medical Centers approved the NBB donor program. All postmortem tissue has been collected from donors for whom a written informed consent for a brain autopsy and the use of the material and clinical information for research purposes has been obtained by the NBB.

AcTau174 ultrasensitive assay

Assay setup

The concentration of AcTau174 in samples was quantified using an in-house developed Simoa-based immunoassay, whereby a monoclonal antibody raised against acetylated lysine 174 (mouse anti-AcTau174 IgG2A, clone 1B8C, provided by the Gan Laboratory, Weill Cornell Medicine) was used as the capture antibody coupled to paramagnetic beads and a monoclonal antibody raised against the N-terminal region of tau (ADx NeuroSciences, ADx204) was used as the biotinylated detector antibody. In vitro acetylated recombinant full-length tau-441 (provided by the Gan Laboratory, Weill Cornell Medicine) was used as the assay calibrator. Calibrators and samples were diluted using assay diluent (Quanterix, 101556). Development, optimization and validation of the assay were performed at the Neurochemistry Laboratory of Amsterdam University Medical Centers, Vrije Universiteit, the Netherlands.

Analytical validation

Analytical validation was conducted according to international consensus guidelines⁸¹. Assay precision, lower limit of detection (LLOD), dilutional linearity and parallelism were assessed. For assay precision, three samples with low, medium and high AcTau174 concentrations were measured over three separate runs on different days, and used to calculate assay repeatability and intermediate precision. The repeatability CV for the low, medium and high samples were 12.7%, 7.8% and 6.1%, respectively. The intermediate precision CV for the low, medium and high samples were 19.4%, 14.9% and 10.7%, respectively. Specificity of the 1B8C antibody for acetylated lysine residue 174, in comparison to acetylated lysine residue 274 and nonacetylated tau, was previously demonstrated⁸² (antibody referred to as clone 2). Further specificity of the 1B8C antibody and potential interference of adjacent post-translational modifications on antibody binding were assessed by testing five C-terminally biotinylated synthetic peptides in the tau region surrounding residue 174 (amino acid range of 158–185; ADx NeuroSciences). These peptides included the following modifications: (1) acetylation at K174; (2) acetylation at K174 and phosphorylation at T175; (3) acetylation at K163; (4) acetylation at K180; and (5) an unmodified peptide without acetylation or phosphorylation. Each peptide was first immobilized on a 96-well streptavidin-coated plate (Corning, 3590) at a concentration of 10 ng ml⁻¹ for 1 h. After washing the plate, the 1B8C antibody was added at serially diluted concentrations (0.05, 0.1, 0.5, 1, 5, 10 and 100 ng ml⁻¹) and incubated for 1 h. After another washing step, the plate was incubated with peroxidase-labeled goat antimouse IgG (Jackson ImmunoResearch, 115-035-044) at a 1:5,000 dilution for 30 min. After a final wash, TMB substrate solution was added to each well and allowed to develop for 30 min. The reaction was stopped with 1 M sulfuric acid, and absorbance was measured at 450 nm with a reference at 600 nm using a microplate reader. The buffer for all dilutions and washing steps was 0.1% casein buffer with 0.1% Tween-20, and all incubations occurred at room temperature.

Measurement of AcTau174 in clinical samples

AcTau174 measurements were performed in duplicate using the in-house developed assay on the Simoa HD-X analyzer. Samples were stored at −80 °C as per international biobanking guidelines⁸³, and before analysis, thawed at room temperature followed by centrifugation at 10,000g for 10 min. For analysis, CSF samples were diluted twofold and brain tissue lysates were diluted 50,000-fold into assay diluent. Brain tissue lysate values were normalized by expressing the concentration as AcTau174 per mg total protein. Concentrations that were measured below the LLOD of the assay were excluded for analysis (20 CSF samples from 11 controls, 4 presymptomatic FTD, 2 FTLD-Tau, 2 MCI-AD and 1 DLB). Median CV of duplicate sample measurements was 18.9% for the CSF samples and 4.7% for the tissue lysates. The inter-run CV was assessed using three internal quality control (QC) samples measured in duplicate at the start and end of each run. Inter-run average CV across seven runs was 9.5% for the high QC, 15.5% for the medium QC and 11.5% for the low QC.

Analysis of postmortem material

Immunohistochemistry

A total of 4% formalin-fixed paraffin-embedded tissue blocks of the middle frontal gyrus, inferior parietal lobule, middle temporal gyrus and hippocampus, all from the right hemisphere, were sliced with a microtome to 5-μm thick sections. These sections were subsequently transferred onto microscope slides (Superfrost Plus, VWR Avantor) and dehydrated overnight at 37 °C to ensure adhesion. After proper adhesion, the material was deparaffinised and rehydrated using the following protocol: submersion thrice in xylene for 5 min for the deparaffinization, followed by the rehydration process of submersion twice in 100% alcohol, once in 96% alcohol and thrice in 70% alcohol, each for 5 min. Thereafter, the sections underwent endogenous peroxidase blocking for 30 min in PBS containing 0.3% H₂O₂, followed by washing the material thrice for 5 min in PBS to rinse off the blocking agent. Next, the antigen retrieval was conducted by immersing the sections in sodium citrate buffer (10 mM sodium citrate, 5 M NaOH, dH2O, pH 6.0) and heating them in an autoclave for 20 min at 120 °C. Once the sections cooled to room temperature, the material was incubated overnight with the primary antibody (mouse anti-AcTau174 IgG2A, 1:9,000 dilution, clone 1B8C, provided by the Gan Laboratory, Weill Cornell Medicine) or (mouse antiphosphorylated tau IgG1, 1:800 dilution, AT8; Thermo Fisher Scientific, MN1020). The following day the sections were washed thrice for 5 min in PBS and incubated for 30 min with EnVision antimouse/rabbit horseradish peroxidase (DAKO, 5007). After incubation, the tissue was washed thrice for 5 min in PBS before incubating in diaminobenzidine (DAB) for 5 min. The sections were then placed under running tap water before undergoing nuclear staining with Mayer’s hematoxylin (Sigma-Aldrich, 517-28-2) for 1 min. Finally, the sections were left under running tap water to enhance the nuclear staining.

After completion of the immunostaining, the sections were dehydrated by reversing the rehydration protocol. The material was submerged thrice in 70% alcohol, twice in 96% alcohol, thrice in 100% alcohol and thrice in xylene, each for a duration of 5 min. Thereafter, DePex mounting medium (SERVA, 18243.02) was used to cover the slides, followed by the securing of the samples with glass coverslips (VWR Radnor, 631-0146). To ensure adhesion, the samples were left to dry overnight at 37 °C.

Quantitative assessment of immunostainings

After scanning the slides using an Olympus VS200 slide scanner, the images were analyzed using QuPath (version 0.5.1-x64). The regions of interest (ROIs) in the frontal, temporal and parietal sections were the gray matter, and the ROI in the hippocampus was the CA1-CA2. The outcome measure was the percentage of DAB-stained area per marker, or immunoreactivity. To calculate this, a pixel classifier was trained on a diverse set of selected regions from 5% of the total images. Once the positive and negative regions had been annotated, the parameters for the pixel classifier were set. For the 1B8C antibody, the classifier was set to random trees with a high resolution (1.09 µm px⁻¹) for the DAB channel on a 1.0 scale with a Gaussian feature. For the AT8 antibody, the classifier was set to random trees with a high resolution (1.09 µm px⁻¹) for the DAB channel on a 1.0 scale with a Gaussian and weighted deviation feature. These parameters were subsequently applied to the previously selected ROIs, with the minimum display brightness set to 0.00, maximum display set to 255.0 and γ set to 1.

Tissue homogenates

Brain tissue homogenates were prepared from the frontal region of the left hemisphere. At autopsy, tissue was snap frozen in liquid nitrogen and stored in cryovials at −80 °C. Using a cryostat, frozen tissue was cut into 20 × 20 µm sections and stored in 2 ml low protein-binding microcentrifuge tubes on dry ice. Lysis buffer was prepared by combining Mammalian Protein Extraction Reagent (Thermo Fisher Scientific, 78501) with protease inhibitors (25×; Roche, 11873580001) and phosphoSTOP (10×; Roche, 04906845001). Ice-cold lysis buffer was added to tissue samples (0.1 g ml⁻¹) on ice for 30 min, followed by vortexing of samples. Following this, samples were centrifuged for 10 min at 11,000g, 4 °C, the supernatant was pipetted to another microcentrifuge tube and the pellet was stored on dry ice. The same volume of lysis buffer was added to resuspend the pellet, incubated on ice for 30 min and vortexed. The sample was then centrifuged a second time for 10 min at 11,000g, 4 °C and the supernatants were pooled, aliquoted and stored at −80 °C until analysis. The total protein concentration in each homogenate was measured using a BCA Protein Assay Kit (Thermo Fisher Scientific, 23225).

Statistical analysis

All statistical analyses were performed using R (version 4.3.2). A two-sided P < 0.05 was considered statistically significant. Biomarker concentrations were not normally distributed; therefore, natural logarithm-transformed values or nonparametric statistics were used. Group differences in demographic and clinical variables were assessed using chi-squared tests for categorical variables or Kruskal–Wallis tests for continuous variables, with Bonferroni multiple comparisons correction. (Sub)group differences in log-transformed CSF AcTau174 were evaluated using linear regression with age and sex as covariates. To account for disease severity, models additionally included MMSE in the total cohort, or FTLD-CDR SOB in FTLD cases. Post hoc pairwise comparisons were conducted using estimated marginal means with Bonferroni adjustment for multiple testing. Spearman’s correlation was used to evaluate associations between untransformed CSF AcTau174 and measures of clinical disease severity, untransformed fluid biomarker concentrations or age-adjusted w scores for CSF and serum NfL^84,85. For individuals with follow-up neuropsychological testing data available, the relationship between baseline CSF AcTau174, or other fluid biomarkers, with changes in MMSE or FTLD-CDR over time was evaluated using linear mixed-effects models with a random intercept and random slope. To ensure comparability of effect sizes among biomarkers in this analysis, log-transformed biomarker concentrations were converted to z scores across the entire dataset. The linear mixed-effects models were adjusted for age, sex, years of education, follow-up time in years and an interaction between biomarker and time. Associations between baseline CSF AcTau174 or other biomarkers (z scored to compare effect sizes) with risk of death during follow-up were evaluated using Cox proportional hazards models, adjusted for age and sex. The time to event was coded as the time between lumbar puncture and death if an individual died during follow-up, or the total follow-up time if an individual was alive at the last follow-up visit. Receiver operating curve analysis determined the ability of fluid biomarkers to differentiate among selected groups, and DeLong’s test was used to compare the AUC between two models. Per brain region, group differences in log-transformed AcTau174 immunoreactivity were evaluated using linear regression with age at death and sex as covariates. Post hoc pairwise comparisons were conducted using estimated marginal means with Bonferroni adjustment for multiple testing. Associations between AcTau174 immunoreactivity in the hippocampus, frontal, parietal and temporal cortex were evaluated using Spearman’s correlation. Per brain region, associations between log-transformed AcTau174 immunoreactivity and neuropathological changes were assessed using linear regression, adjusted for age at death and sex, which included an interaction term between AcTau174 immunoreactivity and pathological diagnosis. Per brain region, associations between log-transformed AcTau174 immunoreactivity and CSF AcTau174 concentrations were assessed using linear regression. The interval between lumbar puncture and tissue collection at death was evaluated as a potential confounder and included in the model if its addition changed the regression coefficient of CSF AcTau174 by ≥10%. Associations between AcTau174 immunoreactivity in the frontal cortex and AcTau174/total protein concentrations in brain homogenates were evaluated using Spearman’s correlation.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Online content

Any methods, additional references, Nature Portfolio reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-026-04341-6.

Supplementary information

Supplementary Information^{(568.8KB, pdf)}

Supplementary Tables 1–5 and Figs. 1–5.

Reporting Summary^{(2.7MB, pdf)}

Peer Review File^{(200.7KB, pdf)}

Supplementary Data 1^{(11KB, xlsx)}

Supporting data for Supplementary Fig. 2: Exact ρ and P values for Spearman’s correlation analyses between CSF AcTau174 and neuropsychological or fluid biomarker data in controls (n = 109), DLB (n = 31), MCI-AD (n = 97), AD (n = 145), FTLD-TDP (n = 52), FTLD-Tau (n = 79) and svPPA (n = 32) in the ADC.

Acknowledgements

We would like to thank all participants of the studies for their contribution to this research. Research of the Alzheimer Center Amsterdam is part of the neurodegeneration research program of Amsterdam Neuroscience. Alzheimer Center Amsterdam is supported by Stichting Alzheimer Nederland and Stichting Steun Alzheimercentrum Amsterdam. The SPIN cohort received funding from the Fondo de Investigaciones Sanitario, Instituto de Salud Carlos III (PI21/00791, PI14/01126, PI17/01019, PI20/01473, PI13/01532, PI16/01825, PI18/00335, PI19/00882, PI18/00435, PI21/01395, PI22/00611, PI22/00307, INT19/00016, INT23/00048, PI17/01896, AC19/00103, PI21/00063, PI24/00968, PI24/01087 and DTS22/00111) and the CIBERNED program (Program 1, AD to A.L.), jointly supported by Fondo Europeo de Desarrollo Regional, Unión Europea, ‘Una manera de hacer Europa’. The SPIN cohort was also supported by the National Institutes of Health (NIA grants 1R01AG056850-01A1, R21AG056974 and R01AG061566), by Generalitat de Catalunya (2017-SGR-547, SLT006/17/125, SLT006/17/119, SLT002/16/408), ‘Marató TV3’ foundation grants (20141210, 044412 and 20142610), a grant from the Fundació Bancaria La Caixa to RB (DABNI project), Fundació Catalana Síndrome de Down and Fundació Víctor Grífols i Lucas. Horizon 21 Consortium is partly funded by the Jérôme Lejeune Foundation. The UPenn cohort received research support from the US National Institutes on Aging (P01-AG066597, P30-AG-072979 and R01-AG087258). M.I.J.H. is a recipient of ABOARD, which is a public–private partnership receiving funding from ZonMW (73305095007) and Health~Holland, Topsector Life Sciences & Health (PPP allowance; LSHM20106). More than 30 partners participate in ABOARD (www.aboard-project.nl), one of which is the Dutch Brain Foundation (Hersenstichting). Y.S.H. is supported with a PPP allowance from Health~Holland. H.S. was supported by ZonMW Onderzoeksprogramma Dementie (TAP-Dementia project 10510032120003 to W.M.v.d.F.), and Alzheimer Nederland (WE.03-2022-07 to H.S. and C.E.T.), Dioraphte Foundation (grant 09-02-03-00), Association for Frontotemporal Dementias Research Grant 2009, Netherlands Organization for Scientific Research (grant HCMI 056-13-018), ZonMw Memorabel (Deltaplan Dementie, project 733 051 042), Joint Programme-Neurodegenerative Disease (JPND) Research-GENFI-PROX (733051109), ZonMW Onderzoeksprogramma Dementie (YOD-INCLUDED, project 10510032120002), EU JPND Research-GENFI-PROX, Alzheimer Nederland (WE.09-2022-05), Gieskes-Strijbis Fonds and the Bluefield Project. D.A. acknowledges support from the Institute of Health Carlos III (ISCIII), Spain (PI18/00435, PI22/00611, INT19/00016 and INT23/00048), jointly funded by Fondo Europeo de Desarrollo Regional, Unión Europea, ‘Una manera de hacer Europa’ and by the Department of Health Generalitat de Catalunya PERIS program (SLT006/17/125). I.I.-G. acknowledges support from ISCIII, Spain (PI21/00791 and PI24/00598), jointly funded by Fondo Europeo de Desarrollo Regional, Unión Europea, ‘Una manera de hacer Europa’. I.I.-G. is a senior Atlantic Fellow for Equity in Brain Health at the Global Brain Health Institute (GBHI) and receives funding from the Alzheimer’s Association (AACSF-21-850193) and the Alzheimer Society (GBHI, ALZ UK-21-72097). I.I.-G. was also supported by the Juan Rodés Contract (JR20/0018) from the Carlos III National Institute of Health of Spain, partly funded by the European Social Fund. A.L.B. has received research support from the US National Institutes on Aging—NIH (U19AG063911, R01AG078457, R01AG073482, R56AG075744, R01AG038791, RF1AG077557, P01AG019724, R01AG071756 and U24AG057437); Rainwater Charitable Foundation, Bluefield Project to Cure FTD, GHR Foundation, Alzheimer’s Association, Association for Frontotemporal Degeneration, Gates Ventures, Alzheimer’s Drug Discovery Foundation, UCSF Parkinson’s Spectrum Disorders Center and the University of California Cures AD Program. S.C.M.d.B. is a recipient of the Early Career Grant from Alzheimer Nederland. Research of C.E.T is supported by the European Commission (Marie Curie International Training Network, grant agreements 860197 (MIRIADE) and 101119596 (TAME)), Innovative Medicines Initiatives 3TR (Horizon 2020, grant 831434) EPND (IMI 2 Joint Undertaking (JU), grant 101034344) and JPND (bPRIDE, CCAD), European Partnership on Metrology, cofinanced from the European Union’s Horizon Europe Research and Innovation Programme and by the Participating States (22HLT07 NEuroBioStand), Horizon Europe (PREDICTFTD, 101156175), CANTATE project funded by the Alzheimer Drug Discovery Foundation, Alzheimer Association, Michael J Fox Foundation, Health Holland, the Dutch Research Council (ZonMW), Alzheimer Drug Discovery Foundation, The Selfridges Group Foundation, Alzheimer Netherlands. C.E.T. is a recipient of ABOARD, which is a public–private partnership receiving funding from ZonMW (73305095007) and Health~Holland, Topsector Life Sciences & Health (PPP allowance; LSHM20106). C.E.T. is also a recipient of TAP-dementia, a ZonMW-funded project (10510032120003) under the Dutch National Dementia Strategy. This study was funded by the Alzheimer’s Drug Discovery Foundation, which had no role in the study design; the collection, analysis or interpretation of data; the writing of the report; or the decision to submit the paper for publication.

Extended data

Author contributions

M.I.J.H., Y.S.H., E.v.d.E., I.M.W.V., L.G., Y.P. and C.E.T. conceptualized the study. M.I.J.H. performed the formal analysis, developed the methodology and visualized the study. M.I.J.H., B.v.d.W. and L.V. performed the investigation. M.I.J.H., K.A.Q.C., E.S., S.C.M.d.B., F.H.D., J.F.H.d.H., T.H.J.M., D.A., D.J.I., J.J.M.H., W.M.v.d.F., H.S., L.G., Y.P. and C.E.T. conducted resource management. E.v.d.E. and C.E.T. were responsible for funding acquisition. M.I.J.H., Y.S.H., I.M.W.V. and C.E.T. contributed to writing the original draft of the paper. All authors contributed to writing, reviewing and editing of the final version of the paper.

Peer review

Peer review information

Nature Medicine thanks Nicolas Barthelemy, Simon Ducharme and Marc Suárez-Calvet for their contribution to the peer review of this work. Peer reviewer reports are available. Primary Handling Editor: Jerome Staal, in collaboration with the Nature Medicine team.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable research request, which will be reviewed by the research team from the Amsterdam UMC, typically within 2 weeks. Upon positive evaluation without obligations regarding intellectual property or confidentiality, data will be shared through Data Transfer Agreements.

Competing interests

M.I.J.H is listed as an inventor on a patent application concerning the use of AcTau174 as a biomarker (EPO application EP26151547.2). L.G. is the founder and consultant of Aeton Therapeutics and is listed as an inventor on a patent application concerning the AcTau antibody used in this study. J.J.M.H. is currently employed by F. Hoffmann-La Roche AG. H.S. is a member of the European Reference Network for Rare Neurological Diseases—Project ID 739510. I.M.W.V. receives speaker honorarium from Quanterix and is a consultant to Neurogen Biomarking. I.M.W.V. also participates in a collaborative research project with Quanterix and Olink, supported by PPP allowance from Health~Holland, and additional collaborative projects with Nitrase Therapeutics, Roche Diagnostics and Sysmex. All payments are made directly to her institution. E.H.T. is a cofounder and Chief Scientific Officer of Neurogen Biomarking. She is listed as an inventor on a patent application concerning the use of AcTau174 as a biomarker (EPO application EP26151547.2). D.J.I. reports NIH grants and industry funding to his institution for clinical trials by Alector, CervoMed, Denali, Passage Bio and Prevail. Funding sources are P01-AG066597 and P30-AG-072979. E.B.L. has received consulting fees from Eli Lilly. D.A. has participated in advisory boards for Fujirebio-Europe, Roche Diagnostics, Grifols S.A. and Lilly, and has received speaker honoraria from Fujirebio-Europe, Roche Diagnostics, Nutricia, Krka Farmacéutica S.L., Zambon S.A.U., Neuraxpharm, Alter Medica, Lilly and Esteve Pharmaceuticals S.A. D.A. also holds a patent for markers of synaptopathy in neurodegenerative diseases (licensed to ADx NeuroSciences N.V., WO2019175379). J.F. reported receiving personal fees for service on the advisory boards, adjudication committees or speaker honoraria from Alnylam, AC Immune, Adamed, Alzheon, Biogen, Eisai, Esteve, Fujirebio, Ionis, Laboratorios Carnot, Life Molecular Imaging, Lilly, Novo Nordisk, Perha, Roche, Zambon and outside the submitted work. He reports holding a patent for markers of synaptopathy in neurodegenerative disease (licensed to ADx, EPI8382175.0). A.L. participated in advisory board meetings from Almirall, Beckman-Coulter, Biogen, Eisai, Fujirebio-Europe, Grifols, Lilly, Novartis, Novo Nordisk, Nutricia, Otsuka Pharmaceutical, Roche and Zambon, and received speaker honoraria from Almirall, Esteve, Lilly, Biogen, KRKA, Nutricia and Zambon. A.L. is co-author of a patent on antibodies for amyloid precursor, methods and uses thereof European priority (EP25382226). I.I.-G. participated in advisory boards from UCB and Nutricia, and received speaker honoraria from Almirall, Esteve Pharmaceuticals S.A., Kern Pharma, Krka Farmacéutica S.L., Lilly, Nutricia and Zambon S.A.U. A.L.B. served as a paid consultant to Alector, Alexion, Arrowhead, Arvinas, Eli Lilly, Janssen, Merck, Neurocrine, Novartis, Oligomerix, Ono, Oscotec, Switch and Transposon. He is a scientific cofounder of Neurovanda. His institution received research support from Biogen and Eisai for serving as a site investigator for clinical trials, as well as from Regeneron. E.S. is an employee of ADx NeuroSciences, Gent, Belgium, which is a partner in ABOARD (www.aboard-project.nl). As of 1 November, 2025, W.M.v.d.F. is executive director at Alzheimer Nederland, Amersfoort, the Netherlands. Before 1 November 2025, research programs of W.M.v.d.F. have been supported by ZonMW, NWO, EU JPND, EU-IHI, Alzheimer Nederland, Hersenstichting CardioVascular Onderzoek Nederland, Health~Holland, Topsector Life Sciences & Health, stichting Dioraphte, Noaber foundation, Pieter Houbolt Fonds, Gieskes-Strijbis Fonds, stichting Equilibrio, Edwin Bouw fonds, Pasman stichting, Philips, Biogen MA, Novartis-NL, Life-MI, AVID, Roche BV, Eli Lilly-NL, Fujifilm, Eisai, Combinostics. W.M.v.d.F. is a recipient of ABOARD, which is a public–private partnership receiving funding from ZonMW (73305095007) and Health~Holland, Topsector Life Sciences & Health (PPP allowance; LSHM20106). Before 1 November 2025, W.M.v.d.F. served as an invited speaker for Biogen MA, Danone, Eisai, WebMD Neurology (Medscape), Novo Nordisk, Springer Healthcare and European Brain Council. W.M.v.d.F has been a consultant to Oxford Health Policy Forum CIC, Roche, Biogen MA, Eisai, Eli Lilly, Owkin France and Nationale Nederlanden Ventures. W.M.v.d.F has participated in advisory boards of Biogen MA, Roche and Eli Lilly. In 2024–2025, W.M.v.d.F has been a member of the steering committee of phase 3 EVOKE/EVOKE+ studies (Novo Nordisk). In 2025, W.M.v.d.F has been a member of the steering committee op phase 3 Trontinemab study (Roche). All funding has been paid to Amsterdam UMC. W.M.v.d.F was associate editor of Alzheimer’s Research & Therapy in 2020/2021. W.M.v.d.F was associate editor at Brain 2021–2025. W.M.v.d.F is chair of the Scientific Leadership Group of InRAD. W.M.v.d.F is also a member of the supervisory board (Raad van Toezicht) Trimbos Instituut. C.E.T. is listed as an inventor on a patent application concerning the use of AcTau174 as a biomarker (EPO application EP26151547.2). C.E.T. has ‘research contracts’ with Acumen, ADx Neurosciences, AC Immune, Alamar, Aribio, Axon Neurosciences, Beckman-Coulter, BioConnect, Bioorchestra, Brainstorm Therapeutics, C2N diagnostics, Celgene, Cognition Therapeutics, EIP Pharma, Eisai, Eli Lilly, Fujirebio, Instant Nano Biosensors, Merck, Muna, Nitrase Therapeutics, Novo Nordisk, Olink, PeopleBio, Quanterix, Roche, Sysmex, Toyama, Vaccinex, Vivoryon. She is editor-in-chief of Alzheimer Research and Therapy, and serves on editorial boards of Molecular Neurodegeneration, Alzheimer’s & Dementia, Neurology—Neuroimmunology & Neuroinflammation, Medidact Neurologie/Springer, and is a committee member to define guidelines for Cognitive disturbances, and one for acute Neurology in the Netherlands. She has ‘consultancy/speaker contracts’ for Aribio, Biogen, Beckman-Coulter, Cognition Therapeutics, Danaher, Eisai, Eli Lilly, Janssen, Merck, Neurogen Biomarking, Nordic Biosciences, Novo Nordisk, Novartis, Olink, Quanterix, Roche, Sanofi and Veravas. The other authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

is available for this paper at 10.1038/s41591-026-04341-6.

Supplementary information

The online version contains supplementary material available at 10.1038/s41591-026-04341-6.

References

1.Gorno-Tempini, M. L. et al. Classification of primary progressive aphasia and its variants. Neurology76, 1006–1014 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Rascovsky, K. et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain134, 2456–2477 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Rademakers, R., Neumann, M. & Mackenzie, I. R. Advances in understanding the molecular basis of frontotemporal dementia. Nat. Rev. Neurol.8, 423–434 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Bang, J., Spina, S. & Miller, B. L. Frontotemporal dementia. Lancet386, 1672–1682 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Seelaar, H., Rohrer, J. D., Pijnenburg, Y. A., Fox, N. C. & van Swieten, J. C. Clinical, genetic and pathological heterogeneity of frontotemporal dementia: a review. J. Neurol. Neurosurg. Psychiatry82, 476–486 (2011). [DOI] [PubMed] [Google Scholar]
6.Greaves, C. V. & Rohrer, J. D. An update on genetic frontotemporal dementia. J. Neurol.266, 2075–2086 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Irwin, D. Preparing for the age of therapeutic trials in frontotemporal lobar degeneration. J. Neurol. Neurosurg. Psychiatry93, 115 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Boxer, A. L. et al. New directions in clinical trials for frontotemporal lobar degeneration: methods and outcome measures. Alzheimers Dement.16, 131–143 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Suárez-Calvet, M. et al. Plasma phosphorylated TDP-43 levels are elevated in patients with frontotemporal dementia carrying a C9orf72 repeat expansion or a GRN mutation. J. Neurol. Neurosurg. Psychiatry85, 684–691 (2014). [DOI] [PubMed] [Google Scholar]
10.Katisko, K. et al. Serum total TDP-43 levels are decreased in frontotemporal dementia patients with C9orf72 repeat expansion or concomitant motoneuron disease phenotype. Alzheimers Res. Ther.14, 151 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Ren, Y. et al. TDP-43 and phosphorylated TDP-43 levels in paired plasma and CSF samples in amyotrophic lateral sclerosis. Front. Neurol.12, 663637 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
12.Grossman, M. et al. Phosphorylated tau as a candidate biomarker for amyotrophic lateral sclerosis. JAMA Neurol.71, 442–448 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Irwin, D. J. et al. Ante mortem cerebrospinal fluid tau levels correlate with postmortem tau pathology in frontotemporal lobar degeneration. Ann. Neurol.82, 247–258 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Lleó, A. et al. A 2-step cerebrospinal algorithm for the selection of frontotemporal lobar degeneration subtypes. JAMA Neurol.75, 738–745 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Abu-Rumeileh, S. et al. Cerebrospinal fluid biomarkers in patients with frontotemporal dementia spectrum: a single-center study. J. Alzheimers Dis.66, 551–563 (2018). [DOI] [PubMed] [Google Scholar]
16.Borroni, B. et al. CSF p-tau181/tau ratio as biomarker for TDP pathology in frontotemporal dementia. Amyotroph. Lateral Scler. Frontotemporal Degener.16, 86–91 (2015). [DOI] [PubMed] [Google Scholar]
17.Meeter, L. H. H. et al. Clinical value of neurofilament and phospho-tau/tau ratio in the frontotemporal dementia spectrum. Neurology90, e1231–e1239 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Cousins, K. A. Q. et al. Distinguishing frontotemporal lobar degeneration tau from TDP-43 using plasma biomarkers. JAMA Neurol.79, 1155–1164 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Pijnenburg, Y. A., Verwey, N. A., van der Flier, W. M., Scheltens, P. & Teunissen, C. E. Discriminative and prognostic potential of cerebrospinal fluid phosphoTau/tau ratio and neurofilaments for frontotemporal dementia subtypes. Alzheimers Dement. (Amst.)1, 505–512 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Feneberg, E., Gray, E., Ansorge, O., Talbot, K. & Turner, M. R. Towards a TDP-43-based biomarker for ALS and FTLD. Mol. Neurobiol.55, 7789–7801 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Kuiperij, H. B. et al. Tau rather than TDP-43 proteins are potential cerebrospinal fluid biomarkers for frontotemporal lobar degeneration subtypes: a pilot study. J. Alzheimers Dis.55, 585–595 (2017). [DOI] [PubMed] [Google Scholar]
22.Chatterjee, M. et al. Plasma extracellular vesicle tau and TDP-43 as diagnostic biomarkers in FTD and ALS. Nat. Med.30, 1771–1783 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
23.Irwin, K. E. et al. A fluid biomarker reveals loss of TDP-43 splicing repression in presymptomatic ALS-FTD. Nat. Med.30, 382–393 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Scialò, C. et al. TDP-43 real-time quaking induced conversion reaction optimization and detection of seeding activity in CSF of amyotrophic lateral sclerosis and frontotemporal dementia patients. Brain Commun.2, fcaa142 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Fontana, E. et al. Detection of TDP-43 seeding activity in the olfactory mucosa from patients with frontotemporal dementia. Alzheimers Dement.20, 1156–1165 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Del Campo, M. et al. New developments of biofluid-based biomarkers for routine diagnosis and disease trajectories in frontotemporal dementia. Alzheimers Dement.18, 2292–2307 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Cohen, T. J. et al. The acetylation of tau inhibits its function and promotes pathological tau aggregation. Nat. Commun.2, 252 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Tracy, T. E. et al. Acetylated tau obstructs KIBRA-mediated signaling in synaptic plasticity and promotes tauopathy-related memory loss. Neuron90, 245–260 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Lucke-Wold, B. et al. Role of tau acetylation in Alzheimer’s disease and chronic traumatic encephalopathy: the way forward for successful treatment. J. Neurol. Neurosurg.4, 140 (2017). [PMC free article] [PubMed] [Google Scholar]
30.Min, S. W. et al. Acetylation of tau inhibits its degradation and contributes to tauopathy. Neuron67, 953–966 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Zhang, S. et al. Acetylation of lysine 82 initiates TDP-43 nuclear loss of function by disrupting its nuclear import. Preprint at bioRxiv10.1101/2024.09.04.611121 (2024).
32.Ko, Y. H. et al. Single acetylation-mimetic mutation in TDP-43 nuclear localization signal disrupts importin α1/β signaling. J. Mol. Biol.436, 168751 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Cohen, T. J. et al. An acetylation switch controls TDP-43 function and aggregation propensity. Nat. Commun.6, 5845 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Garcia Morato, J. et al. Sirtuin-1 sensitive lysine-136 acetylation drives phase separation and pathological aggregation of TDP-43. Nat. Commun.13, 1223 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.García Morato, J., Gloeckner, C. J. & Kahle, P. J. Proteomics elucidating physiological and pathological functions of TDP-43. Proteomics23, e2200410 (2023). [DOI] [PubMed] [Google Scholar]
36.Min, S. W. et al. Critical role of acetylation in tau-mediated neurodegeneration and cognitive deficits. Nat. Med.21, 1154–1162 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.Caballero, B. et al. Acetylated tau inhibits chaperone-mediated autophagy and promotes tau pathology propagation in mice. Nat. Commun.12, 2238 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Montine, T. J. et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease: a practical approach. Acta Neuropathol.123, 1–11 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Liu, E. et al. Accuracy of blood-based neurofilament light to different genetic frontotemporal dementia from primary psychiatric disorders. J. Alzheimers Dis.106, 1337–1354 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Hok, A. H. Y. S. et al. Large-scale CSF proteome profiling identifies biomarkers for accurate diagnosis of frontotemporal dementia. Mol. Neurodegener.20, 93 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
41.Robinson, J. L. et al. Primary tau pathology, not copathology, correlates with clinical symptoms in PSP and CBD. J. Neuropathol. Exp. Neurol.79, 296–304 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Sen, T., Saha, P. & Sen, N. Nitrosylation of GAPDH augments pathological tau acetylation upon exposure to amyloid-β. Sci. Signal.11, eaao6765 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Cicognola, C. et al. Novel tau fragments in cerebrospinal fluid: relation to tangle pathology and cognitive decline in Alzheimer’s disease. Acta Neuropathol.137, 279–296 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Kim, Y. C. & Jeong, B. H. In silico evaluation of acetylation mimics in the 27 lysine residues of human tau protein. Curr. Alzheimer Res.16, 379–387 (2019). [DOI] [PubMed] [Google Scholar]
45.Grinberg, L. T. et al. Argyrophilic grain disease differs from other tauopathies by lacking tau acetylation. Acta Neuropathol.125, 581–593 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Irwin, D. J. et al. Acetylated tau neuropathology in sporadic and hereditary tauopathies. Am. J. Pathol.183, 344–351 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
47.Irwin, D. J. et al. Acetylated tau, a novel pathological signature in Alzheimer’s disease and other tauopathies. Brain135, 807–818 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Chakraborty, P. et al. Acetylation discriminates disease-specific tau deposition. Nat. Commun.14, 5919 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
49.Tomé, S. O. et al. TDP-43 pathology is associated with increased tau burdens and seeding. Mol. Neurodegener.18, 71 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Rana, J. K. et al. CTE-type tau filaments in Alzheimer’s disease with co-morbid LATE-NC. Preprint at bioRxiv10.1101/2025.10.21.683735 (2025).
51.Josephs, K. A. et al. C-terminal and full length TDP-43 specie differ according to FTLD-TDP lesion type but not genetic mutation. Acta Neuropathol. Commun.7, 100 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Fiesel, F. C. et al. Knockdown of transactive response DNA-binding protein (TDP-43) downregulates histone deacetylase 6. EMBO J.29, 209–221 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.Trzeciakiewicz, H. et al. An HDAC6-dependent surveillance mechanism suppresses tau-mediated neurodegeneration and cognitive decline. Nat. Commun.11, 5522 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
54.Neumann, M. et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science314, 130–133 (2006). [DOI] [PubMed] [Google Scholar]
55.Van der Flier, W. M. et al. Optimizing patient care and research: the Amsterdam Dementia Cohort. J. Alzheimers Dis.41, 313–327 (2014). [DOI] [PubMed] [Google Scholar]
56.Slot, R. E. R. et al. Subjective Cognitive Impairment Cohort (SCIENCe): study design and first results. Alzheimers Res. Ther.10, 76 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
57.Willemse, E. A. J. et al. Diagnostic performance of Elecsys immunoassays for cerebrospinal fluid Alzheimer’s disease biomarkers in a nonacademic, multicenter memory clinic cohort: the ABIDE project. Alzheimers Dement. (Amst.)10, 563–572 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Strong, M. J. et al. Amyotrophic lateral sclerosis - frontotemporal spectrum disorder (ALS-FTSD): revised diagnostic criteria. Amyotroph. Lateral Scler. Frontotemporal Degener.18, 153–174 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Armstrong, M. J. et al. Criteria for the diagnosis of corticobasal degeneration. Neurology80, 496–503 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Höglinger, G. U. et al. Clinical diagnosis of progressive supranuclear palsy: the movement disorder society criteria. Mov. Disord.32, 853–864 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.McKhann, G. M. et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement.7, 263–269 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.McKeith, I. G. et al. Diagnosis and management of dementia with Lewy bodies: fourth consensus report of the DLB Consortium. Neurology89, 88–100 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Petersen, R. C. et al. Mild cognitive impairment: a concept in evolution. J. Intern. Med.275, 214–228 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
64.Krudop, W. A. et al. Building a new paradigm for the early recognition of behavioral variant frontotemporal dementia: late onset frontal lobe syndrome study. Am. J. Geriatr. Psychiatry22, 735–740 (2014). [DOI] [PubMed] [Google Scholar]
65.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5 5th edn (American Psychiatric Publishing, 2013).
66.Knopman, D. S. et al. Development of methodology for conducting clinical trials in frontotemporal lobar degeneration. Brain131, 2957–2968 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
67.Verhage, F. Intelligentie en leeftijd: Onderzoek bij Nederlanders van twaalf tot zevenenzeventig jaar[in Dutch] (Van Gorcum, 1964).
68.Spinelli, E. G. et al. Typical and atypical pathology in primary progressive aphasia variants. Ann. Neurol.81, 430–443 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
69.Mackenzie, I. R. The neuropathology of FTD associated with ALS. Alzheimer Dis. Assoc. Disord.21, S44–S49 (2007). [DOI] [PubMed] [Google Scholar]
70.Rösler, T. W. et al. Four-repeat tauopathies. Prog. Neurobiol.180, 101644 (2019). [DOI] [PubMed] [Google Scholar]
71.Dopper, E. G. et al. Structural and functional brain connectivity in presymptomatic familial frontotemporal dementia. Neurology83, e19–e26 (2014). [DOI] [PubMed] [Google Scholar]
72.Papma, J. M. et al. Cognition and gray and white matter characteristics of presymptomatic C9orf72 repeat expansion. Neurology89, 1256–1264 (2017). [DOI] [PubMed] [Google Scholar]
73.Ludolph, A. et al. A revision of the El Escorial criteria—2015. Amyotroph. Lateral Scler. Frontotemporal Degener.16, 291–292 (2015). [DOI] [PubMed] [Google Scholar]
74.Toledo, J. B. et al. A platform for discovery: the University of Pennsylvania Integrated Neurodegenerative Disease Biobank. Alzheimers Dement.10, 477–484 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
75.Alcolea, D. et al. The Sant Pau Initiative on Neurodegeneration (SPIN) cohort: a data set for biomarker discovery and validation in neurodegenerative disorders. Alzheimers Dement. (N.Y.)5, 597–609 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
76.Alcolea, D. et al. Agreement of amyloid PET and CSF biomarkers for Alzheimer’s disease on Lumipulse. Ann. Clin. Transl. Neurol.6, 1815–1824 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
77.Klioueva, N. M. et al. BrainNet Europe’s code of conduct for brain banking. J. Neural Transm. (Vienna)122, 937–940 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
78.Mackenzie, I. R. & Neumann, M. Molecular neuropathology of frontotemporal dementia: insights into disease mechanisms from postmortem studies. J. Neurochem.138, 54–70 (2016). [DOI] [PubMed] [Google Scholar]
79.Hyman, B. T. et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement.8, 1–13 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
80.Fujishiro, H. et al. Validation of the neuropathologic criteria of the third consortium for dementia with Lewy bodies for prospectively diagnosed cases. J. Neuropathol. Exp. Neurol.67, 649–656 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
81.Andreasson, U. et al. A practical guide to immunoassay method validation. Front. Neurol.6, 179 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
82.Parra Bravo, C. et al. Anti-acetylated-tau immunotherapy is neuroprotective in tauopathy and brain injury. Mol. Neurodegener.19, 51 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
83.Teunissen, C. E. et al. A consensus protocol for the standardization of cerebrospinal fluid collection and biobanking. Neurology73, 1914–1922 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
84.Vermunt, L. et al. Age- and disease-specific reference values for neurofilament light presented in an online interactive support interface. Ann. Clin. Transl. Neurol.9, 1832–1837 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
85.Dijkstra, J. I. R. et al. TREM2 risk variants and associated endophenotypes in Alzheimer’s disease. Alzheimers Res. Ther.17, 57 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Information^{(568.8KB, pdf)}

Supplementary Tables 1–5 and Figs. 1–5.

Reporting Summary^{(2.7MB, pdf)}

Peer Review File^{(200.7KB, pdf)}

Supplementary Data 1^{(11KB, xlsx)}

Data Availability Statement

[CR1] 1.Gorno-Tempini, M. L. et al. Classification of primary progressive aphasia and its variants. Neurology76, 1006–1014 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR2] 2.Rascovsky, K. et al. Sensitivity of revised diagnostic criteria for the behavioural variant of frontotemporal dementia. Brain134, 2456–2477 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR3] 3.Rademakers, R., Neumann, M. & Mackenzie, I. R. Advances in understanding the molecular basis of frontotemporal dementia. Nat. Rev. Neurol.8, 423–434 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR4] 4.Bang, J., Spina, S. & Miller, B. L. Frontotemporal dementia. Lancet386, 1672–1682 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR5] 5.Seelaar, H., Rohrer, J. D., Pijnenburg, Y. A., Fox, N. C. & van Swieten, J. C. Clinical, genetic and pathological heterogeneity of frontotemporal dementia: a review. J. Neurol. Neurosurg. Psychiatry82, 476–486 (2011). [DOI] [PubMed] [Google Scholar]

[CR6] 6.Greaves, C. V. & Rohrer, J. D. An update on genetic frontotemporal dementia. J. Neurol.266, 2075–2086 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR7] 7.Irwin, D. Preparing for the age of therapeutic trials in frontotemporal lobar degeneration. J. Neurol. Neurosurg. Psychiatry93, 115 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR8] 8.Boxer, A. L. et al. New directions in clinical trials for frontotemporal lobar degeneration: methods and outcome measures. Alzheimers Dement.16, 131–143 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR9] 9.Suárez-Calvet, M. et al. Plasma phosphorylated TDP-43 levels are elevated in patients with frontotemporal dementia carrying a C9orf72 repeat expansion or a GRN mutation. J. Neurol. Neurosurg. Psychiatry85, 684–691 (2014). [DOI] [PubMed] [Google Scholar]

[CR10] 10.Katisko, K. et al. Serum total TDP-43 levels are decreased in frontotemporal dementia patients with C9orf72 repeat expansion or concomitant motoneuron disease phenotype. Alzheimers Res. Ther.14, 151 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR11] 11.Ren, Y. et al. TDP-43 and phosphorylated TDP-43 levels in paired plasma and CSF samples in amyotrophic lateral sclerosis. Front. Neurol.12, 663637 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR12] 12.Grossman, M. et al. Phosphorylated tau as a candidate biomarker for amyotrophic lateral sclerosis. JAMA Neurol.71, 442–448 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR13] 13.Irwin, D. J. et al. Ante mortem cerebrospinal fluid tau levels correlate with postmortem tau pathology in frontotemporal lobar degeneration. Ann. Neurol.82, 247–258 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR14] 14.Lleó, A. et al. A 2-step cerebrospinal algorithm for the selection of frontotemporal lobar degeneration subtypes. JAMA Neurol.75, 738–745 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR15] 15.Abu-Rumeileh, S. et al. Cerebrospinal fluid biomarkers in patients with frontotemporal dementia spectrum: a single-center study. J. Alzheimers Dis.66, 551–563 (2018). [DOI] [PubMed] [Google Scholar]

[CR16] 16.Borroni, B. et al. CSF p-tau181/tau ratio as biomarker for TDP pathology in frontotemporal dementia. Amyotroph. Lateral Scler. Frontotemporal Degener.16, 86–91 (2015). [DOI] [PubMed] [Google Scholar]

[CR17] 17.Meeter, L. H. H. et al. Clinical value of neurofilament and phospho-tau/tau ratio in the frontotemporal dementia spectrum. Neurology90, e1231–e1239 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR18] 18.Cousins, K. A. Q. et al. Distinguishing frontotemporal lobar degeneration tau from TDP-43 using plasma biomarkers. JAMA Neurol.79, 1155–1164 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR19] 19.Pijnenburg, Y. A., Verwey, N. A., van der Flier, W. M., Scheltens, P. & Teunissen, C. E. Discriminative and prognostic potential of cerebrospinal fluid phosphoTau/tau ratio and neurofilaments for frontotemporal dementia subtypes. Alzheimers Dement. (Amst.)1, 505–512 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR20] 20.Feneberg, E., Gray, E., Ansorge, O., Talbot, K. & Turner, M. R. Towards a TDP-43-based biomarker for ALS and FTLD. Mol. Neurobiol.55, 7789–7801 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR21] 21.Kuiperij, H. B. et al. Tau rather than TDP-43 proteins are potential cerebrospinal fluid biomarkers for frontotemporal lobar degeneration subtypes: a pilot study. J. Alzheimers Dis.55, 585–595 (2017). [DOI] [PubMed] [Google Scholar]

[CR22] 22.Chatterjee, M. et al. Plasma extracellular vesicle tau and TDP-43 as diagnostic biomarkers in FTD and ALS. Nat. Med.30, 1771–1783 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR23] 23.Irwin, K. E. et al. A fluid biomarker reveals loss of TDP-43 splicing repression in presymptomatic ALS-FTD. Nat. Med.30, 382–393 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR24] 24.Scialò, C. et al. TDP-43 real-time quaking induced conversion reaction optimization and detection of seeding activity in CSF of amyotrophic lateral sclerosis and frontotemporal dementia patients. Brain Commun.2, fcaa142 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR25] 25.Fontana, E. et al. Detection of TDP-43 seeding activity in the olfactory mucosa from patients with frontotemporal dementia. Alzheimers Dement.20, 1156–1165 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR26] 26.Del Campo, M. et al. New developments of biofluid-based biomarkers for routine diagnosis and disease trajectories in frontotemporal dementia. Alzheimers Dement.18, 2292–2307 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR27] 27.Cohen, T. J. et al. The acetylation of tau inhibits its function and promotes pathological tau aggregation. Nat. Commun.2, 252 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR28] 28.Tracy, T. E. et al. Acetylated tau obstructs KIBRA-mediated signaling in synaptic plasticity and promotes tauopathy-related memory loss. Neuron90, 245–260 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR29] 29.Lucke-Wold, B. et al. Role of tau acetylation in Alzheimer’s disease and chronic traumatic encephalopathy: the way forward for successful treatment. J. Neurol. Neurosurg.4, 140 (2017). [PMC free article] [PubMed] [Google Scholar]

[CR30] 30.Min, S. W. et al. Acetylation of tau inhibits its degradation and contributes to tauopathy. Neuron67, 953–966 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR31] 31.Zhang, S. et al. Acetylation of lysine 82 initiates TDP-43 nuclear loss of function by disrupting its nuclear import. Preprint at bioRxiv10.1101/2024.09.04.611121 (2024).

[CR32] 32.Ko, Y. H. et al. Single acetylation-mimetic mutation in TDP-43 nuclear localization signal disrupts importin α1/β signaling. J. Mol. Biol.436, 168751 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR33] 33.Cohen, T. J. et al. An acetylation switch controls TDP-43 function and aggregation propensity. Nat. Commun.6, 5845 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR34] 34.Garcia Morato, J. et al. Sirtuin-1 sensitive lysine-136 acetylation drives phase separation and pathological aggregation of TDP-43. Nat. Commun.13, 1223 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR35] 35.García Morato, J., Gloeckner, C. J. & Kahle, P. J. Proteomics elucidating physiological and pathological functions of TDP-43. Proteomics23, e2200410 (2023). [DOI] [PubMed] [Google Scholar]

[CR36] 36.Min, S. W. et al. Critical role of acetylation in tau-mediated neurodegeneration and cognitive deficits. Nat. Med.21, 1154–1162 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR37] 37.Caballero, B. et al. Acetylated tau inhibits chaperone-mediated autophagy and promotes tau pathology propagation in mice. Nat. Commun.12, 2238 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR38] 38.Montine, T. J. et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease: a practical approach. Acta Neuropathol.123, 1–11 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR39] 39.Liu, E. et al. Accuracy of blood-based neurofilament light to different genetic frontotemporal dementia from primary psychiatric disorders. J. Alzheimers Dis.106, 1337–1354 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR40] 40.Hok, A. H. Y. S. et al. Large-scale CSF proteome profiling identifies biomarkers for accurate diagnosis of frontotemporal dementia. Mol. Neurodegener.20, 93 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR41] 41.Robinson, J. L. et al. Primary tau pathology, not copathology, correlates with clinical symptoms in PSP and CBD. J. Neuropathol. Exp. Neurol.79, 296–304 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR42] 42.Sen, T., Saha, P. & Sen, N. Nitrosylation of GAPDH augments pathological tau acetylation upon exposure to amyloid-β. Sci. Signal.11, eaao6765 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR43] 43.Cicognola, C. et al. Novel tau fragments in cerebrospinal fluid: relation to tangle pathology and cognitive decline in Alzheimer’s disease. Acta Neuropathol.137, 279–296 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR44] 44.Kim, Y. C. & Jeong, B. H. In silico evaluation of acetylation mimics in the 27 lysine residues of human tau protein. Curr. Alzheimer Res.16, 379–387 (2019). [DOI] [PubMed] [Google Scholar]

[CR45] 45.Grinberg, L. T. et al. Argyrophilic grain disease differs from other tauopathies by lacking tau acetylation. Acta Neuropathol.125, 581–593 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR46] 46.Irwin, D. J. et al. Acetylated tau neuropathology in sporadic and hereditary tauopathies. Am. J. Pathol.183, 344–351 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR47] 47.Irwin, D. J. et al. Acetylated tau, a novel pathological signature in Alzheimer’s disease and other tauopathies. Brain135, 807–818 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR48] 48.Chakraborty, P. et al. Acetylation discriminates disease-specific tau deposition. Nat. Commun.14, 5919 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR49] 49.Tomé, S. O. et al. TDP-43 pathology is associated with increased tau burdens and seeding. Mol. Neurodegener.18, 71 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR50] 50.Rana, J. K. et al. CTE-type tau filaments in Alzheimer’s disease with co-morbid LATE-NC. Preprint at bioRxiv10.1101/2025.10.21.683735 (2025).

[CR51] 51.Josephs, K. A. et al. C-terminal and full length TDP-43 specie differ according to FTLD-TDP lesion type but not genetic mutation. Acta Neuropathol. Commun.7, 100 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR52] 52.Fiesel, F. C. et al. Knockdown of transactive response DNA-binding protein (TDP-43) downregulates histone deacetylase 6. EMBO J.29, 209–221 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR53] 53.Trzeciakiewicz, H. et al. An HDAC6-dependent surveillance mechanism suppresses tau-mediated neurodegeneration and cognitive decline. Nat. Commun.11, 5522 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR54] 54.Neumann, M. et al. Ubiquitinated TDP-43 in frontotemporal lobar degeneration and amyotrophic lateral sclerosis. Science314, 130–133 (2006). [DOI] [PubMed] [Google Scholar]

[CR55] 55.Van der Flier, W. M. et al. Optimizing patient care and research: the Amsterdam Dementia Cohort. J. Alzheimers Dis.41, 313–327 (2014). [DOI] [PubMed] [Google Scholar]

[CR56] 56.Slot, R. E. R. et al. Subjective Cognitive Impairment Cohort (SCIENCe): study design and first results. Alzheimers Res. Ther.10, 76 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR57] 57.Willemse, E. A. J. et al. Diagnostic performance of Elecsys immunoassays for cerebrospinal fluid Alzheimer’s disease biomarkers in a nonacademic, multicenter memory clinic cohort: the ABIDE project. Alzheimers Dement. (Amst.)10, 563–572 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR58] 58.Strong, M. J. et al. Amyotrophic lateral sclerosis - frontotemporal spectrum disorder (ALS-FTSD): revised diagnostic criteria. Amyotroph. Lateral Scler. Frontotemporal Degener.18, 153–174 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR59] 59.Armstrong, M. J. et al. Criteria for the diagnosis of corticobasal degeneration. Neurology80, 496–503 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR60] 60.Höglinger, G. U. et al. Clinical diagnosis of progressive supranuclear palsy: the movement disorder society criteria. Mov. Disord.32, 853–864 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR61] 61.McKhann, G. M. et al. The diagnosis of dementia due to Alzheimer’s disease: recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement.7, 263–269 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR62] 62.McKeith, I. G. et al. Diagnosis and management of dementia with Lewy bodies: fourth consensus report of the DLB Consortium. Neurology89, 88–100 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR63] 63.Petersen, R. C. et al. Mild cognitive impairment: a concept in evolution. J. Intern. Med.275, 214–228 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR64] 64.Krudop, W. A. et al. Building a new paradigm for the early recognition of behavioral variant frontotemporal dementia: late onset frontal lobe syndrome study. Am. J. Geriatr. Psychiatry22, 735–740 (2014). [DOI] [PubMed] [Google Scholar]

[CR65] 65.American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5 5th edn (American Psychiatric Publishing, 2013).

[CR66] 66.Knopman, D. S. et al. Development of methodology for conducting clinical trials in frontotemporal lobar degeneration. Brain131, 2957–2968 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR67] 67.Verhage, F. Intelligentie en leeftijd: Onderzoek bij Nederlanders van twaalf tot zevenenzeventig jaar[in Dutch] (Van Gorcum, 1964).

[CR68] 68.Spinelli, E. G. et al. Typical and atypical pathology in primary progressive aphasia variants. Ann. Neurol.81, 430–443 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR69] 69.Mackenzie, I. R. The neuropathology of FTD associated with ALS. Alzheimer Dis. Assoc. Disord.21, S44–S49 (2007). [DOI] [PubMed] [Google Scholar]

[CR70] 70.Rösler, T. W. et al. Four-repeat tauopathies. Prog. Neurobiol.180, 101644 (2019). [DOI] [PubMed] [Google Scholar]

[CR71] 71.Dopper, E. G. et al. Structural and functional brain connectivity in presymptomatic familial frontotemporal dementia. Neurology83, e19–e26 (2014). [DOI] [PubMed] [Google Scholar]

[CR72] 72.Papma, J. M. et al. Cognition and gray and white matter characteristics of presymptomatic C9orf72 repeat expansion. Neurology89, 1256–1264 (2017). [DOI] [PubMed] [Google Scholar]

[CR73] 73.Ludolph, A. et al. A revision of the El Escorial criteria—2015. Amyotroph. Lateral Scler. Frontotemporal Degener.16, 291–292 (2015). [DOI] [PubMed] [Google Scholar]

[CR74] 74.Toledo, J. B. et al. A platform for discovery: the University of Pennsylvania Integrated Neurodegenerative Disease Biobank. Alzheimers Dement.10, 477–484 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR75] 75.Alcolea, D. et al. The Sant Pau Initiative on Neurodegeneration (SPIN) cohort: a data set for biomarker discovery and validation in neurodegenerative disorders. Alzheimers Dement. (N.Y.)5, 597–609 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR76] 76.Alcolea, D. et al. Agreement of amyloid PET and CSF biomarkers for Alzheimer’s disease on Lumipulse. Ann. Clin. Transl. Neurol.6, 1815–1824 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR77] 77.Klioueva, N. M. et al. BrainNet Europe’s code of conduct for brain banking. J. Neural Transm. (Vienna)122, 937–940 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR78] 78.Mackenzie, I. R. & Neumann, M. Molecular neuropathology of frontotemporal dementia: insights into disease mechanisms from postmortem studies. J. Neurochem.138, 54–70 (2016). [DOI] [PubMed] [Google Scholar]

[CR79] 79.Hyman, B. T. et al. National Institute on Aging-Alzheimer’s Association guidelines for the neuropathologic assessment of Alzheimer’s disease. Alzheimers Dement.8, 1–13 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR80] 80.Fujishiro, H. et al. Validation of the neuropathologic criteria of the third consortium for dementia with Lewy bodies for prospectively diagnosed cases. J. Neuropathol. Exp. Neurol.67, 649–656 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR81] 81.Andreasson, U. et al. A practical guide to immunoassay method validation. Front. Neurol.6, 179 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR82] 82.Parra Bravo, C. et al. Anti-acetylated-tau immunotherapy is neuroprotective in tauopathy and brain injury. Mol. Neurodegener.19, 51 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR83] 83.Teunissen, C. E. et al. A consensus protocol for the standardization of cerebrospinal fluid collection and biobanking. Neurology73, 1914–1922 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR84] 84.Vermunt, L. et al. Age- and disease-specific reference values for neurofilament light presented in an online interactive support interface. Ann. Clin. Transl. Neurol.9, 1832–1837 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[CR85] 85.Dijkstra, J. I. R. et al. TREM2 risk variants and associated endophenotypes in Alzheimer’s disease. Alzheimers Res. Ther.17, 57 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

An acetylated Tau-174 CSF biomarker discriminates between TDP-43 and tau pathology in patients with frontotemporal lobar degeneration

Madison I J Honey

Yanaika S Hok-A-Hin

Elisabeth H Thijssen

Katheryn A Q Cousins

Babette van der Weijden

Lisanne in ‘t Veld

Erik Stoops

Sterre C M de Boer

Flora H Duits

Julie F H de Houwer

Tjado H J Morrema

Daniel Alcolea

Ignacio Illán-Gala

Juan Fortea

Alberto Lleó

Adam L Boxer

David J Irwin

Edward B Lee

Leslie Shaw

Corey T McMillan

David A Wolk

Jeroen J M Hoozemans

Wiesje M van der Flier

Emma van der Ende

Harro Seelaar

Inge M W Verberk

Li Gan

Yolande Pijnenburg

Charlotte E Teunissen

Abstract

Main

Results

AcTau174 prototype assay and analytical validation

Extended Data Table 1.

Extended Data Fig. 1. AcTau174 per clinical disease group in the pilot cohort.

Extended Data Fig. 2. Analytical validation of the CSF AcTau174 assay on the Simoa platform.

Participant characteristics

Table 1.

CSF AcTau174 is elevated in FTLD-TDP compared to all other clinical groups

Fig. 1. CSF AcTau174 across clinical disease groups of the ADC, UPenn, SPIN and FTD-RisC and CSF AcTau174/tTau ratio in the ADC.

Table 2.

Extended Data Fig. 3. CSF AcTau174 measurements in autopsy- or genetically-confirmed FTLD cases of the ADC.

Extended Data Fig. 4. CSF AcTau174 across clinical disease groups of the ADC, excluding FTD cases with AD copathology.

CSF AcTau174 robustly discriminates FTLD-TDP from FTLD-Tau and controls

Fig. 2. Diagnostic accuracy of CSF AcTau174 and other fluid biomarkers in the ADC, SPIN and UPenn cohorts.

Extended Data Fig. 5. CSF and serum biomarker concentrations across clinical disease groups of the ADC.

Extended Data Fig. 6. Diagnostic performance of biofluid biomarkers in differentiating clinical groups of the ADC.

CSF AcTau174 reflects disease severity and neurodegeneration across cohorts

Baseline CSF AcTau174 predicts future cognitive and functional decline

Fig. 3. The effect of baseline AcTau174 on longitudinal changes of MMSE and FTLD-CDR scores.

Extended Data Table 2.

Extended Data Table 3.

AcTau174 in brain tissue suggests diagnostic relevance of soluble tau species

Extended Data Table 4.

Fig. 4. AcTau174 in brain tissue.

Discussion

Methods

Patient cohorts

ADC

FTD-RisC

UPenn cohort

SPIN cohort

The Netherlands Brain Bank postmortem cohort

Ethical approval

AcTau174 ultrasensitive assay

Assay setup

Analytical validation

Measurement of AcTau174 in clinical samples

Analysis of postmortem material

Immunohistochemistry

Quantitative assessment of immunostainings

Tissue homogenates

Statistical analysis

Reporting summary

Online content

Supplementary information

Acknowledgements

Extended data