Plasma N‐terminal tau fragment is an amyloid‐dependent biomarker in Alzheimer's disease

Abstract

INTRODUCTION
Novel fluid biomarkers for tracking neurodegeneration specific to Alzheimer's disease (AD) are greatly needed.

METHODS
Using two independent well‐characterized cohorts (n = 881 in total), we investigated the group differences in plasma N‐terminal tau (NT1‐tau) fragments across different AD stages and their association with cross‐sectional and longitudinal amyloid beta (Aβ) plaques, tau tangles, brain atrophy, and cognitive decline.

RESULTS
Plasma NT1‐tau significantly increased in symptomatic AD and displayed positive associations with Aβ PET (positron emission tomography) and tau PET. Higher baseline NT1‐tau levels predicted greater tau PET, with 2‐ to 10‐year intervals and faster longitudinal Aβ PET increases, AD‐typical neurodegeneration, and cognitive decline. Plasma NT1‐tau showed negative correlations with baseline regional brain volume and thickness, superior to plasma brain‐derived tau (BD‐tau) and neurofilament light (NfL) in Aβ‐positive participants.

DISCUSSION
This study suggests that plasma NT1‐tau is an Aβ‐dependent biomarker and outperforms BD‐tau and NfL in detecting cross‐sectional neurodegeneration in the AD continuum.

Highlights
Plasma N‐terminal tau (NT1‐tau) was specifically increased in the A+/T+ stage.
Plasma NT1‐tau was positively associated with greater amyloid beta (Aβ) and tau PET (positron emission tomography) accumulations.
Higher plasma NT1‐tau predicted greater tau burden and faster Aβ increases.
Plasma NT1‐tau was more related to neurodegeneration than plasma brain‐derived tau (BD‐tau) and neurofilament light (NfL).

Full text

Received: 10 October 2024 Revised: 20 December 2024 Accepted: 23 December 2024
DOI: 10.1002/alz.14550
RESEARCH ARTICLE
Plasma N-terminal tau fragment is an amyloid-dependent
biomarker in Alzheimer’s disease
Guoyu Lan1Laihong Zhang1,2Anqi Li1,3Wenqing Ran4Jieqin Lv5
Fernando Gonzalez-Ortiz6,7Yue Cai1Pan Sun1Lin Liu1Jie Yang1,8
Zhengbo He1Lili Fang1Xin Zhou1Yalin Zhu1,3Zhen Liu1Xuhui Chen9
Xiang Fan10 Dai Shi11 Chenghui Ye11 Linsen Xu12 Qingyong Wang13
Kaj Blennow6,7,14,15,16 Guanxun Cheng10 for the Alzheimer’s Disease Neuroimaging
Initiative Pengcheng Ran5Lu Wang4Tengfei Guo1,17,18
1Institute of Neurological and Psychiatric Disorders, Shenzhen Bay Laboratory, Shenzhen, China
2School of Biology and Biological Engineering, South China University of Technology,Guangzhou, China
3Division of Life Science, The Hong Kong University of Science and Technology, Hong Kong, China
4Department of Nuclear Medicine, The First Affiliated Hospital of Jinan University, Guangzhou, China
5Department of Nuclear Medicine, Guangdong Provincial Hospital of Chinese Medicine, Guangzhou University of Chinese Medicine, Guangzhou, China
6Institute of Neuroscience and Physiology, University of Gothenburg, Mölndal, Sweden
7Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölndal, Sweden
8Department of Neurology, Xuanwu Hospital, Capital Medical University, Beijing, China
9Department of Neurology, Peking University Shenzhen Hospital, Shenzhen, China
10Department of Medical Imaging, Peking University Shenzhen Hospital, Shenzhen, China
11Neurology Medicine Center, The Seventh Affiliated Hospital, Sun Yat-sen University, Shenzhen, China
12Department of Medical Imaging, Shenzhen Guangming District People’s Hospital, Shenzhen, China
13Department of Neurology, Shenzhen Guangming District People’s Hospital, Shenzhen, China
14Paris Brain Institute, ICM, Pitié-Salpêtrière Hospital, Sorbonne University, Paris, France
15Neurodegenerative Disorder Research Center, Division of Life Sciences and Medicine, Hefei, China
16Department of Neurology, Institute on Aging and Brain Disorders, University of Science and Technology of China and First Affiliated Hospital of USTC,Hefei, China
17Institute of Biomedical Engineering, Shenzhen Bay Laboratory, Shenzhen, China
18Peking University Shenzhen Graduate School, Peking University, Shenzhen, China
Correspondence
Dr. Tengfei Guo, Institute of Neurological and
Psychiatric Disorders, Shenzhen Bay
Laboratory, No.5 KelianRoad, Shenzhen,
518132, China.
Email: tengfei.guo@szbl.ac.cn
Abstract
INTRODUCTION: Novel fluid biomarkers for tracking neurodegeneration specific to
Alzheimer’s disease (AD) are greatly needed.
METHODS: Using two independent well-characterized cohorts (n=881 in total),
we investigated the group differences in plasma N-terminal tau (NT1-tau) fragments
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited and is not used for commercial purposes.
© 2025 The Author(s). Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association.
Alzheimer’s Dement. 2025;21:e14550. wileyonlinelibrary.com/journal/alz 1of14
https://doi.org/10.1002/alz.14550

2of14 LAN ET AL.
Dr. Lu Wang,Department of Nuclear Medicine,
The First Affiliated Hospital of Jinan
University, Guangzhou 510630, China.
Email: l_wang1009@foxmail.com
Dr. Pengcheng Ran, Department of Nuclear
Medicine, Guangdong Provincial Hospital of
Chinese Medicine, Guangzhou University of
Chinese Medicine, Guangzhou, 510641, China.
Email: cmurpc@163.com
Data used in preparation of this article were
obtained from the Alzheimer’s Disease
Neuroimaging Initiative (ADNI) database
(adni.loni.usc.edu). As such, the investigators
within the ADNI contributed to the design and
implementation of ADNI and/or provided data
but did not participate in the analysis or
writing of this report. A complete listing of
ADNI investigators can be found
at: http://adni.loni.usc.edu/wp-
content/uploads/how_to_apply/
ADNI_Acknowledgement_List.pdf.
Funding information
National Natural Science Foundation of China,
Grant/AwardNumbers: 82171197,
82301380; Guangdong Basic and Applied
Basic Science Foundation for Distinguished
Young Scholars, Grant/Award Number:
2023B1515020113; Shenzhen Science and
TechnologyProgram, Grant/Award Number:
RCYX20221008092935096; National Key
Research and Development Program of China,
Grant/AwardNumber: 2023YFC3605400;
Shenzhen Bay Laboratory,Grant/Award
Number: S241101004-1; Lingang Laboratory,
Grant/AwardNumber:
LG-GG-202401-ADA070600
across different AD stages and their association with cross-sectional and longitudinal
amyloid beta (Aβ) plaques, tau tangles, brain atrophy, and cognitive decline.
RESULTS: Plasma NT1-tau significantly increased in symptomatic AD and displayed
positive associations with AβPET (positron emission tomography) and tau PET. Higher
baseline NT1-tau levels predicted greater tau PET, with 2- to 10-year intervals and
faster longitudinal AβPET increases, AD-typical neurodegeneration, and cognitive
decline. Plasma NT1-tau showed negative correlations with baseline regional brain vol-
ume and thickness, superior to plasma brain-derived tau (BD-tau) and neurofilament
light (NfL) in Aβ-positive participants.
DISCUSSION: This study suggests that plasma NT1-tau is an Aβ-dependent biomarker
and outperforms BD-tau and NfL in detecting cross-sectional neurodegeneration in
the AD continuum.
KEYWORDS
Alzheimer’s disease, BD-tau, cognitive decline, neurodegeneration, NfL, NT1-tau
Highlights
∙Plasma N-terminal tau (NT1-tau) was specifically increased in the A+/T+stage.
∙Plasma NT1-tau was positively associated with greater amyloid beta (Aβ)andtau
PET (positron emission tomography) accumulations.
∙Higher plasma NT1-tau predicted greater tau burden and faster Aβincreases.
∙Plasma NT1-tau was more related to neurodegeneration than plasma brain-derived
tau (BD-tau) and neurofilament light (NfL).
1BACKGROUND
Alzheimer’s disease (AD) is neuropathologically characterized by the
abnormal deposition of amyloid beta (Aβ) plaques and neurofibrillary
tangles, which can be quantified by positron emission tomography
(PET) or cerebrospinal fluid (CSF) biomarkers.1–4 With the applica-
tions of anti-Aβimmunotherapies in clinical practice, scalable and
cost-effective biomarker strategies to identify patients at high risk of
AD are warranted. According to the revised Alzheimer’s Association
(AA) framework,5AD is a biological process defined by the appear-
ance of neuropathological features, including the early-changing Core
1 biomarkers and later-changing Core 2 biomarkers. Specific Core 1
biomarkers comprise AβPET (A) and biofluid phosphorylated tau (p-
tau) species (T1) that become abnormal at a time close to AβPET
(i.e., p-tau181, p-tau217, and p-tau231).6–8 These T1 biomarkers are
strongly associated with Aβplaques, which may indicate early tau
pathologyinreactiontoAβpathology.9In contrast, the microtubule-
binding region of tau containing the residue 243 (MTBR-tau243) and
non-phosphorylated mid-region tau (MR-tau) fragments, which seem
to become abnormal later in the AD trajectory,are more closely related
to AD-type tau tangles and thus categorized as Core 2 biomarkers.10,11
In addition, neurodegeneration (N) is a nonspecific but important step
in the AD pathogenic progress. Neurofilament light (NfL) is the most
promising biofluid marker of neurodegeneration and axonal injury
across several diseases, but it cannot distinguish AD from other neu-
rodegenerative disorders.12,13 Regarding total tau, the large overlap
between AD and control groups and close association with p-tau pro-
pose some concerns about its usage as an N biomarker.5,14,15 In this
respect, an urgent need is for new fluid biomarkers to detect and track
AD-type neurodegeneration.
Given the multiple limitations of high-cost PET and invasive CSF
tests in clinical practices, accurate blood biomarkers16 may provide
alternative approaches to detect AD pathology. Blood-based biomark-
ers for prescreening individuals at high risk of AD, combined with fur-
ther confirmation with PET or CSF biomarkers, would be an important
strategy for future AD diagnosis and monitoring.13,17 Technological
development in mass spectrometry and ultrasensitive immunoassays
have yielded promising blood biomarkers to track brainAD pathophys-
iology,while most advances in this field target phosphorylated and non-
phosphorylated variants of tau protein. Previous studies have reported
a large spectrum of tau fragments in plasma, showing that different tau
variants may provide distinct information on AD pathology.18–21 For

LAN ET AL.3of14
instance, plasma N-terminal–targeted tau fragments (NTA-tau) were
observed to be increased across the AD continuum and closely associ-
ated with AD-type tau tangle accumulation.18,22 A recently developed
assay referred to as brain-derived tau (BD-tau) showed higher plasma
levels in identified AD than in non-AD dementia and control groups
and could predict future neurodegeneration and cognitive decline.19,23
Thus, plasma BD-tau is regarded as a potential neurodegeneration
biomarker specific to AD, in contrast to plasma NfL. Previously, Chen
et al. designed an in-house N-terminal tau (NT1-tau) assay targeting
longer NT1-tau fragments containing the mid-region.21 Observational
studies have reported that plasma NT1-tau elevation was related to
increased risk for progressing to AD but not non-AD dementia,21,24
and was predictive of future cognitive decline and brain atrophy bet-
ter than plasma NfL.25 These findings suggest that plasma NT1-tau is
a strong predictor of AD clinically relevant aspects of neurodegenera-
tion and cognitive decline. However, it is unclear how plasma NT1-tau
corresponds with brain AD pathophysiology and how plasma NT1-tau
compares to BD-tau and NfL as a biomarker for detecting and tracking
neurodegeneration in AD.
In this study, we intended to compare the changes of plasma
NT1-tau across different clinical and biological stages of AD in two
independent cohorts. We further assessed the association of plasma
NT1-tau with cross-sectional and longitudinal changes in cerebral Aβ
and tau accumulations, brain volume, and cognition. In addition, we
compared the properties of plasma NT1-tau, BD-tau, and NfL in pre-
dicting brain atrophy and cognitive decline. We aim to systematically
characterize the features of plasma NT1-tau in relationship with AD
pathological changes and to determine whether plasma NT1-tau could
be a biomarker for AD-type neurodegeneration.
2METHODS
2.1 Participants
This study included participants from two independent cohorts: the
Greater-Bay-Area Healthy Aging Brain Study (GHABS) from China
(NCT06183658) and the Alzheimer’s Disease Neuroimaging Initiative
(ADNI) from the United States (NCT00106899). The GHABS cohort
was launched by the Shenzhen Bay Laboratory in May 2021, aiming
to track the pathological features and progression of AD and iden-
tify novel biomarkers for diagnosing AD at an early stage. The GHABS
cohort would recruit individuals between 55 and 90 years of age in
the community and nursing homes, including cognitively normal (CU)
individuals and patients with mild cognitive impairment (MCI) and AD
dementia, following the standard protocol of the ADNI cohort. Further
inclusion, exclusion, and clinical diagnosis criteria have been described
elsewhere.26 The GHABS cohort was approved by the ethical commit-
tees at the Shenzhen Bay Laboratory, and all participants signed the
written informed consent and underwent annual blood sample collec-
tion and cognitive assessments. A subset of participants underwent
baseline and longitudinal AβPET, tau PET, and magnetic resonance
imaging (MRI) scans every 2 years.
RESEARCH IN CONTEXT
1. Systematic review: Literature reviews in PubMed and
Google Scholar suggest that plasma N-terminal tau (NT1-
tau) is linked to an increased risk of Alzheimer’s disease
(AD) and can predict neurodegeneration and cognitive
decline. However, it is unclear how it corresponds with
brain AD pathophysiology. No studies have compared
the performance of plasma NT1-tau, brain-derived tau
(BD-tau), and neurofilament light (NfL) as AD-type neu-
rodegeneration biomarkers.
2. Interpretation: We found that plasma NT1-tau signifi-
cantly increased in symptomatic AD and displayed pos-
itive associations with amyloid beta (Aβ), and tau PET
(positron emission tomography). Elevated baseline NT1-
tau was strongly correlated with greater tau PET burden
and faster AβPET increases, AD-typical neurodegen-
eration, and cognitive decline. Plasma NT1-tau showed
stronger correlations with brain atrophy and cortical
thinning than BD-tau and NfL in Aβ+participants. This
study suggests that plasma NT1-tau is an AD-specific
neurodegeneration biomarker.
3. Future directions: These findings need further validation
using independent cohorts with larger sample sizes and
longer follow-up duration.
The ADNI cohort is known as a multicenter public–private
partnership, led by principal investigator Michael W. Weiner, MD
(ida.loni.usc.edu). The main objective of the ADNI is to determine
whether serial MRI, PET, other biological markers, and clinical and
neuropsychological assessments can be combined to track the pro-
gression of MCI and early AD. The samples used for the current study
were collected from participants between March 7, 2011 and June 10,
2021. All participants were assessed with plasma NT1-tau and at least
one AβPET scan. This study was approved by the institutional review
boards (IRBs) of all participating centers.
2.2 Plasma and CSF biomarker measurements
Plasma NT1-tau concentrations were measured using an in-house
assay on the fully automated Simoa HD-X Analyzer (Quanterix
Corp., Lexington, MA), as described previously.27 Briefly, the NT1-tau
assay used a bead-conjugated capture antibody BT2 at 1.8 mg/mL
(MN1010, ThermoFisher) and a biotinylated detector antibody Tau12
at 0.6 µg/mL (MAB2241, EMD-Millipore). Plasma samples were cen-
trifuged at 14,000 ×gfor 4 min, followed by a 4-fold dilution
with Tau 2.0 sample diluent reagent (Quanterix). The NT1-tau assay
used a three-step protocol at room temperature: (1) Mix stan-
dard/blank/sample and antibody-coated beads for 30 min with a

4of14 LAN ET AL.
washing step. (2) Incubate beads with biotinylated detector antibody
for 10 min with a further washing step. (3) Incubate beads with
150 pM streptavidin-β-galactosidase with a further washing step, then
add enzyme substrate (resorufin β-D-galactopyranoside). The bead-
bearing complexes were resuspended and loaded into Simoa arrays.
The average enzyme unit per bead (AEB) was determined as described
previously.24 Standard curves of AEB versus tau441 concentration
were fitted to a five-parameter logistic function with 1/Y2 weighting.
In the GHABS cohort, plasma concentrations of Aβ42, Aβ40, glial
fibrillary acidic protein (GFAP), NfL, and p-tau181 were quantified by
the Simoa HD-X Analyzer using the commercial NEURO 4-PLEX E and
pTau-181 V2 Advantage kits, following the protocol of the manufac-
turer. Plasma concentrations of p-tau217, p-tau231, and BD-tau were
measured using in-house Simoa assays developed by the University of
Gothenburg.19,28,29 Plasma soluble triggering receptor expressed on
myeloid cells 2 (sTREM2) was determined using a Meso Scale Discov-
ery (MSD) platform–based electro-chemiluminescence assay devel-
oped by the Haass group.30,31 All plasma biomarker measurements
were blinded to clinical diagnosis at the Shenzhen Bay Laboratory
(Shenzhen, China). The optimal cutoff value for plasma p-tau217 was
defined at 5.4 pg/mL after comparing measures of AβPET negative
(Aβ–) CU participants (n=219) with AβPET positive (Aβ+) cognitively
impaired (CI) participants (n=75), as described in Figure S1A.
In the ADNI cohort, CSF Aβ42/Aβ40 was examined by two-
dimensional ultra performance liquid chromatography-tandem mass
spectrometry (2D-UPLC-MS/MS), and CSF total tau and p-tau181
were quantified using the fully automated Roche Elecsys at the Uni-
versity of Pennsylvania. CSF sTREM2 and progranulin were measured
using the in-house MSD assay (Haass group) at Ludwig Maximilian Uni-
versity of Munich.30,32 CSF growth-associated protein 43 (GAP-43)
was measured by an in-house enzyme-linked immunosorbent assay
(ELISA) as described previously.33,34 The optimal cutoff value for CSF
p-tau181 was defined at 23 pg/mL after comparing measures of Aβ–
CU participants (n=313) with Aβ+CI participants (n=421), as
described in Figure S1B.
2.3 Neuroimaging acquisition and analysis
Details on descriptions of structure MRI and PET imaging procedures
in the GHABS and ADNI cohorts have been reported previously26
and given elsewhere (http://adni-info.org), respectively. Hippocam-
pal volume (HCV) and 68 FreeSurfer-defined gray matter volumes
(GMVs) and cortical thicknesses were calculated from the structural
MRI scan via FreeSurfer (V7.2.0).35 The estimated intracranial volume
was used to adjust bilateral HCV, and then the residual HCV was cal-
culated as the difference between the raw and expected HCV as we
described previously.36,37 Cortical thickness in AD-signature atrophy
brain regions (temporal-MetaROI) was computed as a surface-area
weighted average of the mean cortical thickness in the entorhinal,
fusiform, inferior temporal, and middle temporal cortices.16
AβPET scans were performed with [18F]-D3FSP (FSP) or [18 F]-
florbetapir (FBP) tracer in the GHABS cohort and with FBP tracer in
the ADNI cohort. A composite standardized uptake value ratio (SUVR)
of AD summary cortical regions (frontal, cingulate, parietal, and tem-
poral regions) was calculated to evaluate cortical Aβplaques, with the
mean FSP uptake in the brainstem or the mean FBP uptakein the whole
cerebellum as reference regions.38,39 AβPET positive was defined
as composite FSP SUVR ≥0.78 or composite FBP SUVR ≥1.11.40
Slopes of FBP AβPET across 68 pre-defined brain regions were cal-
culated for all the participants with longitudinal AβPET data in the
ADNI cohort, using linear mixed-effects (LME) models (lme4 pack-
age), adjusting for age and sex, and including a random time slope and
intercept.
Tau PET scans were acquired with [18F]-flortaucipir (FTP) tracer in
both cohorts. Cortical tau tangles were evaluated by the SUVR in AD
temporal-MetaROI (entorhinal cortex, parahippocampal gyrus, amyg-
dala, and inferior temporal and middle temporal), with the mean FTP
uptake in the inferior cerebellar cortex as the reference region.41 Tau
PET positive was defined as temporal-MetaROI FTP SUVR ≥1.27.4
2.4 Cognitive assessments
Montreal Cognitive Assessment (MoCA) was used to measure cogni-
tion in the GHABS cohort. LME models were used to calculate slopes
of MoCA for all the participants with longitudinal MoCA data in the
GHABS cohort, adjusting for age, sex, and education and including
a random time slope and intercept. Cognition in the ADNI cohort
was represented by composite scores for memory (ADNI-MEM) and
executive function (ADNI-EF), which were computed by combining
z-scores of several cognitive tests.42,43 Briefly, ADNI-MEM included
the Rey Auditory Verbal Learning Test (two versions), AD Assessment
Schedule-Cognition (three versions), Logical Memory, and Mini-Mental
State Examination (MMSE) data. ADNI-EF included Wechsler Adult
Intelligence Scale-Revised (WAIS-R) Digit Symbol Substitution, Digit
Span Backwards, Trails A and B, Category Fluency (animals and
vegetables), and Clock Drawing data.
2.5 Biological stages defined by fluid and imaging
biomarkers
According to Core 1 biomarkers in the recent AA criteria,5all partic-
ipants were stratified into four different A/T groups (A, determined
by AβPET; T, determined by plasma p-tau217 for GHABS or by CSF
p-tau181 for ADNI), using the cutoff values described above.
2.6 Statistical analysis
Statistical analyses were performed using the program R (v4.3.0,
The R Foundation for Statistical Computing) in two cohorts based
on the available data. Demographics and clinical characteristics of
participants were shown as the number (%) or mean (SD). Plasma and
CSF biomarkers were z-standardized before parametric analyses, and
a significance threshold level of p<0.05 was applied unless otherwise
noted in this study.

LAN ET AL.5of14
Generalized linear models (GLMs) were used to assess the associ-
ation of plasma NT1-tau with other plasma biomarkers in the GHABS
cohort and with CSF biomarkers in the ADNI cohort, adjusting for age
and sex. Then, we compared the group differences in plasma NT1-tau
across the clinical and biological stages of AD, using GLM models with
age and sex as covariates. The Benjamini–Hochberg approach with a
false discovery rate (FDR) of 0.05 was applied for multiple comparisons
correction.
We assessed the cross-sectional association of plasma NT1-tau
(predictor) with AβPET and tau PET (outcome) in both GHABS and
ADNI cohorts, using independent GLM models with age and sex as
covariates. Among participants with longitudinal AβPET scans in the
ADNI cohort, we further explored the association between plasma
NT1-tau and longitudinal changes in AβPET using LME models with
time ×NT1-tau interaction as a predictor, adjusting for baseline Aβ
PET levels in addition to the above covariates. Random intercept
and time slope were also included in LME models. Furthermore, we
evaluated the predictive effect of plasma NT1-tau on longitudinal
slopes of AβPET across 68 FreeSurfer-defined brain regions with the
Benjamini–Hochberg correction (FDR <0.05).
Next, we studied the performance of plasma NT1-tau, BD-tau,
and NfL as biomarkers for tracking neurodegeneration and cognitive
decline using the GHABS cohort. We first assessed the cross-sectional
association of plasma NT1-tau, BD-tau, and NfL with GMV and cor-
tical thickness across 68 brain regions as well as two AD-typical
neurodegeneration markers (temporal-MetaROIcortical thickness and
residual HCV) using GLM models with age, sex, and whole brain
volume as covariates. For participants with repeated MRI scans at
follow-up, independent LME models were used to evaluate further
the association of plasma NT1-tau, BD-tau, and NfL with longitu-
dinal temporal-MetaROI cortical thinning and hippocampal atrophy.
Time ×plasma biomarker interaction was regarded as a predictor, con-
trolling for age, sex, and baseline level of temporal-MetaROI cortical
thickness or residual HCV. With regard to cognitive impairment, we
evaluated the cross-sectional and longitudinal associations of plasma
NT1-tau, BD-tau, and NfL with MoCA scores using GLM models, con-
trolling for age, sex, and education. Finally, in the ADNI cohort, we
validated the association with longitudinal temporal-MetaROI cortical
thinning, hippocampal atrophy, and cognitive decline (ADNI-MEM and
ADNI-EF) for plasma NT1-tau.
In longitudinal analyses, note that plasma biomarker levels were
divided into two subgroups for illustration purposes only, and all
the plasma biomarkers within the interaction term were modeled
continuously.
3RESULTS
3.1 Participants’ characteristics
Table 1summarizes the clinical characteristics at baseline of included
participants. Age and female proportion were not different between
Aβ–andAβ+groups of the GHABS cohort, but greater in Aβ+par-
ticipants of the ADNI cohort. For the GHABS cohort, a total of 349
participants who underwent AβPET and MRI imaging had concurrent
plasma biomarker and cognitive assessments, including 237 CU (22%
Aβ+), 66 MCI (45% Aβ+), and 46 dementia (83% Aβ+). Of the included
participants, 106 had two to three repeats of MRI scans up to 2.7 years
(mean =1.1, SD =0.5 years); 192 had repeated cognitive assessments
every year up to 2.5 years (mean =1.2, SD =0.4 years). A subset of 141
participants underwent additional FTP tau PET imaging at baseline.
For the ADNI cohort, a total of 532 participants who underwent FBP
AβPET scans had available plasma NT1-tau and CSF p-tau181 mea-
sures within 1 year, including 270 CU (31% Aβ+), 167 MCI (97% Aβ+),
and 95 dementia (88% Aβ+). All participants had baseline and longitu-
dinalMRIscanswithameanof3.8yearsfollow-up(SD=2.8 years)
and cognitive assessments with a mean of 4.2 years follow-up (SD =2.9
years). Of the included participants, 367 had concurrent other CSF
biomarker measurements; 381 had repeated AβPET imaging up to 11.3
years (mean =4.7, SD =2.6 years). A subset of 154 participants under-
went FTP tau PET imaging in the future, with intervals of 2 to 10 years
(mean =5.3, SD =1.4 years).
The associations between plasma NT1-tau levels and other biofluid
markers are presented in Figure S2, with the strongest association with
plasma p-tau181 in the GHABS cohort (standardized β[βstd]=0.65
[95% confidence interval (CI), 0.56–0.74], p<0.001) and with CSF
Aβ42/Aβ40 in the ADNI cohort (βstd =−0.20 [95% CI, −0.30 to −0.10],
p<0.001).
3.2 Comparison of plasma NT1-tau across the
clinical diagnosis and biological stages
In the GHABS cohort, plasma NT1-tau levels were significantly higher
in the Aβ+dementia group than in other groups (p<0.001 for all;
Figure 1A). Plasma NT1-tau levels were also increased in the Aβ+MCI
group compared to the Aβ–CU(p=0.016) and Aβ–CI(p=0.009)
groups. In the ADNI cohort, plasma NT1-tau levels were pronouncedly
higher in the Aβ+MCI (p<0.001) and dementia (p=0.001) groups
when compared to the Aβ– CU group (Figure 1B). No significant differ-
ences were observed between the Aβ–CIandAβ+MCI and dementia
after multiple comparisons correction, although plasma NT1-tau lev-
els seemed slightly higher in the latter two groups. Moreover, the
group differences between the Aβ–CU,Aβ+CU, and Aβ–CIwere
insignificant in either cohort.
Across biological classifications, the A+/T+group showed signifi-
cantly higher plasma NT1-tau levels than the A–/T–, A–/T+,andA+/T–
groups in both the GHABS and ADNI cohorts (Figure 1B, D). No differ-
ences were observed between the A–/T–, A–/T+,andA+/T– groups in
either cohort.
3.3 Association of plasma NT1-tau with Aβand
tau pathologies
Next, we tested the association of plasma NT1-tau with cross-sectional
and longitudinal changes in Aβand tau aggregations using PET imaging.
In the GHABS cohort, higher plasma NT1-tau levels were significantly

6of14 LAN ET AL.
TAB L E 1 Demographics and characteristics of participants.
All participants Aβnegative Aβpositive
GHABS cohort n=349 n=228 n=121
Age, years 67.4 (7.4) 66.9 (6.6) 68.2 (8.6)
Female, no. (%) 218 (62) 137 (60) 81 (67)
Diagnosis, CU/MCI/Dementia 237/66/46 184/36/8 53/30/38
MoCA scoresa23.38 (6.13) 25.34 (3.76) 19.57 (7.84)
Residual HCV,cm3−0.40 (1.02) −0.11 (0.75) −0.93 (1.22)
Temporal-MetaROI cortical thickness, mm 2.70 (0.15) 2.73 (0.10) 2.65 (0.19)
Aβtracers, FSP/FBP 278/71 175/53 103/18
Composite FSP PET, SUVR 1.08 (0.17) 0.99 (0.05) 1.23 (0.18)
Composite FBP PET, SUVR 1.10 (0.23) 0.97 (0.05) 1.39 (0.20)
Temporal-MetaROI FTP PET, SUVRb1.34 (0.40) 1.12 (0.06) 1.59 (0.48)
ADNI cohort n=532 n=202 n=330
Age, years 72.7 (6.6) 71.7 (6.3) 73.4 (6.8)
Female, no. (%) 261 (49) 95 (47) 166 (50)
Diagnosis, CU/MCI/Dementia 270/167/95 186/5/11 84/162/84
ADNI-MEM 0.49 (1.12) 1.20 (0.91) 0.05 (1.01)
ADNI-EF 0.44 (1.02) 0.97 (0.87) 0.12 (0.97)
Residual HCV,cm3−0.51 (0.99) −0.04 (0.86) −0.80 (0.96)
Temporal-MetaROI cortical thickness, mm 2.64 (0.14) 2.69 (0.11) 2.61 (0.15)
Composite FBP PET, SUVR 1.26 (0.26) 1.00 (0.06) 1.42 (0.19)
Temporal-MetaROI FTP PET, SUVRc1.32 (0.35) 1.18 (0.10) 1.49 (0.45)
Intervals of FTP PET, Year 5.3 (1.4) 5.6 (1.6) 5.0 (1.1)
Note: Data are presented as mean (SD) or participant number (percentage). Aβstatus was determined by the positivity of FSP PET or FBP PET.
Abbreviations: Aβ, amyloid-β; CU, cognitively unimpaired; EF, executive functioning; FBP, [18F]-florbetapir; FTP, [18F]-flortaucipir; FSP, [18F]-D3FSP;
HCV, hippocampal volume; MCI, mild cognitive impairment; MEM, memory; MoCA, Montreal Cognitive Assessment; PET, positron emission computed
tomography; SUVR, standard uptake value ratio.
aSix participants missing in GHABS.
bA total of 141 participants had FTP PET in GHABS.
cA total of 154 participants had FTP PET with intervals of 2 to 10 years in ADNI.
associated with more Aβburden, assessed by FSP PET (βstd =0.39
[95% CI, 0.28–0.49], p<0.001) and FBP PET (βstd =0.43 [95% CI,
0.22–0.63], p<0.001), and tau tangle aggregation assessed by FTP
tau PET (βstd =0.62 [95% CI, 0.49–0.75], p<0.001) (Figure 2A, B and
Figure S3A). In particular, only Aβ+participants preserved the pos-
itive associations of plasma NT1-tau with FSP AβPET (βstd =0.42
[95% CI, 0.25–0.60], p<0.001) and FTP tau PET (βstd =0.60 [95% CI,
0.41–0.78], p<0.001) (Table S1).
In the ADNI cohort, we observed a similar positive association
between plasma NT1-tau and FBP AβPET at baseline (βstd =0.15
[95% CI, 0.06–0.23], p<0.001) (Figure 2C). Among some participants
who underwent tau PET imaging in the future, higher baseline NT1-tau
levels were associated with greater FTP tau PET burden (βstd =0.14
[95% CI, 0.02–0.25], p=0.022) (Figure 2D). Furthermore, in a subset
of participants with longitudinal AβPET imaging, higher baseline NT1-
tau levels were associated with faster increases in composite FBP Aβ
PET SUVR (Time ×NT1-tau interaction: βstd =0.002 [95% CI, 0.001–
0.003], p<0.001) (Figure 3A, B). Congruent results were obtained
when the models were restricted to Aβ+participants (Time ×NT1-
tau interaction: βstd =0.002 [95% CI, 0.001–0.004], p<0.001) but
not Aβ–participants(p=0.533) (Figure S3B). We then explored the
association between plasma NT1-tau and longitudinal AβPET changes
across 68 FreeSurfer-defined brain regions. Whole-brain region of
interest (ROI) analyses revealed that the predictive effect of plasma
NT1-tau on longitudinal FBP AβPET increases was significant pri-
marily in the pars triangularis of inferior frontal, middle frontal, and
superior temporal regions (Figure 3C and Tables S2–3).
3.4 Cross-sectional association of plasma
NT1-tau, BD-tau, and NfL with GMV and cortical
thickness in the GHABS cohort
Across all participants, brain-wide ROI analyses showed that plasma
levels of NT1-tau and NfL exhibited stronger correlations with cross-
sectional GMV and cortical thickness compared to plasma BD-tau,

LAN ET AL.7of14
FIGURE 1 Comparison of plasma NT1-tau across the clinical diagnosis and biological stages. Plasma NT1-tau levels in theGHABS cohort by
Aβstatus and clinical diagnosis (A) and A/T stages (B). Plasma NT1-tau levels in the ADNI cohort by Aβstatus and clinical diagnosis (C) and A/T
stages (D). The boxplots depict the median (horizontal bar), IQR (hinges), and 1.5 ×IQR (whiskers). Each point represents an individual, and green
dashed lines represent the median values of the Aβ– CU or A–/T– group. Multiple comparisons were corrected using the Benjamini–Hochberg
method (FDR <0.05). The adjusted p-values of the comparisons are shown at the top of the bar, controlling for age and sex. Aβ, amyloid beta; ADNI,
Alzheimer’s Disease Neuroimaging Initiative; CI, cognitively impaired; CU, cognitively unimpaired; FDR, false discovery rate; GHABS,
Greater-Bay-Area Healthy Aging Brain Study; IQR, interquartile range; MCI, mild cognitive impairment.
which was indicated by higher R values and wider regions for plasma
NT1-tau and NfL (Figure 4A). When the model included Aβ+partic-
ipants, the strength and area of NT1-tau correlations with GMV and
cortical thickness were dramatically increased, markedly outperform-
ingbothBD-tauandNfL(Figure4B). Rare correlations were observed
for plasma BD-tau across the whole brain map in either the whole
cohort or the Aβ+group. This indicates that plasma NT1-tau may be
superior to plasma BD-tau and NfL as a biomarker for detecting cross-
sectional neurodegeneration in patients within the AD continuum.
Subsequently, we explored their association with two characterized
markers of AD-typical neurodegeneration temporal-MetaROI cortical
thickness and residual HCV. We found significant negative associations
of plasma NT1-tau and NfL with cross-sectional temporal-MetaROI
cortical thickness and residual HCV across all participants (p<0.001
for all) (Table S4). However, the associations for plasma NT1-tau were
Aβdependent as shown by the higher plasma NT1-tau levels that
were associated with lower temporal-MetaROI cortical thickness
(βstd =−0.34 [95% CI, −0.51 to −0.18], p<0.001) and residual HCV
(βstd =−0.36 [95% CI, −0.52 to −0.20], p<0.001) in the Aβ+group
but not in the Aβ– group (Figure S4). For plasma NfL, the associations
with residual HCV persisted in both Aβ+(βstd =−0.21 [95% CI, −0.39
to −0.03], p=0.026) and Aβ–(βstd =−0.17 [95% CI, −0.30 to −0.05],
p=0.006) groups, and the association with temporal-MetaROI cortical
thickness was maintained in Aβ+group only (βstd =−0.22 [95% CI,
−0.40 to −0.03], p=0.024). In contrast, plasma BD-tau showed a
significant association with temporal-MetaROI cortical thickness
(p=0.037) and residual HCV (p=0.019) across all participants but not
in either Aβsubgroup.
3.5 Association of plasma NT1-tau, BD-tau, and
NfL with longitudinal neurodegeneration and
cognitive decline
Finally, we investigated whether plasma NT1-tau, BD-tau, and NfL
could predict future AD-typical neurodegeneration and cognitive

8of14 LAN ET AL.
FIGURE 2 Association of plasma NT1-tau with AβPET and tau PET. Association of plasma NT1-tau with baseline FSP AβPET (n=278) (A) and
FTP tau PET (n=141) (B) in the GHABS cohort. Association of plasma NT1-tau with baseline FBP AβPET (n=532) (C) and future FTP tau PET
(n=154) (D) in the ADNI cohort. The points (blue, Aβ–; red, Aβ+) and solid lines represent the participants and regression lines (95% CI). βstd and p
were computed using GLM, controlling for age and sex. Spearman correlation coefficient (rho) was also calculated. ***p<0.001; **p<0.01;
*p<0.05. βstd, Standardized β;Aβ, amyloid beta; ADNI, Alzheimer’s Disease Neuroimaging Initiative; CI, confidence interval; FBP,
[18F]-florbetapir; FSP, [18F]-D3FSP; FTP, [18F]-flortaucipir; GHABS, Greater-Bay-Area Healthy Aging Brain Study; GLM, Generalized linear
models; PET, positron emission tomography; SUVR, standard uptake value ratio.
decline in the GHABS cohort. Both higher baseline levels of NT1-tau
and BD-tau were associated with steeper longitudinal decreases in
temporal-MetaROI cortical thickness (βstd =−0.02 [95% CI, −0.03
to −0.004], p=0.01 for NT1-tau; βstd =−0.02 [95% CI, −0.03 to
−0.01], p=0.002 for BD-tau) and residual HCV (βstd =−0.06 [95%
CI, −0.11 to −0.01], p=0.021 for NT1-tau; βstd =-0.09 [95% CI,
−0.14 to −0.04], p<0.001 for BD-tau) (Figure 5and Table S5). There
was no significant association between plasma NfL and longitudinal
temporal-MetaROI cortical thickness and residual HCV. Congruent
results were obtained when the models were restricted to the Aβ+
group (Table S5). With respect to cognitive function, three plasma
biomarkers were negatively associated with cross-sectional MoCA
scores in the whole cohort (p<0.001 for all) (Table S4). Strati-
fied by Aβstatus, the association for plasma NT1-tau was significant
only in the Aβ+group (βstd =−0.35 [95% CI, −0.49 to −0.20],
p<0.001), but the association for plasma NfL was sustained in both
Aβ+(βstd =−0.32 [95% CI, −0.48 to −0.16], p<0.001) and Aβ–
(βstd =−0.19 [95% CI, −0.31 to −0.08], p<0.001) groups (Figure
S5A–C). Plasma BD-tau did not correlate with baseline MoCA in
either Aβsubgroup. Longitudinally, three plasma biomarkers were
negatively associated with faster rates of declines in MoCA scores
across all participants (p<0.001 for all) (Figure S5D–F), whereas the
association was maintained only for NT1-tau when the models were
restricted to Aβ+participants (βstd =−0.53 [95% CI, −0.80 to −0.27],
p<0.001).
The analyses for plasma NT1-tau were repeated in the ADNI cohort.
Higher baseline NT1-tau levels were associated with faster decreases
in temporal-MetaROI cortical thickness (Time ×NT1-tau interaction:
βstd =−0.003 [95% CI, −0.005 to −0.0002], p=0.033) but not longitu-
dinal residual HCV (p=0.264) (Figure S6). In addition, plasma NT1-tau
also showed a significant association with more pronounced cognitive
decline, assessed by slopes of ADNI-MEM (βstd =−0.14 [95% CI, −0.23
to −0.06], p<0.001) and ADNI-EF (βstd =−0.15 [95% CI, −0.23 to
−0.06], p<0.001; Figure S7).

LAN ET AL.9of14
FIGURE 3 Association of plasma NT1-tau with longitudinal AβPET. Association of plasma NT1-tau with longitudinal changes in composite
FBP AβPET (n=381) in the ADNI cohort (A). LME models were used with composite FBP AβPET SUVR as the outcome and the time ×NT1-tau
interaction as the predictor with random intercepts and time slopes. Age and sex were used as covariates. For illustration purposes only, the
regression lines of longitudinal changes are shown for participants with plasma NT1-tau levels below versus above the median. The trajectory of
longitudinal changes in composite FBP AβPET, stratified by NT1-tau status (B). Brain-wise ROI analyses for the associations between plasma
NT1-tau and longitudinal FBP AβPET across 68 brain regions (C). The threshold was set at p<0.05 after Benjamini–Hochberg corrections.
***p<0.001. Aβ, amyloid beta; ADNI, Alzheimer’s Disease Neuroimaging Initiative; FBP, [18F]-florbetapir; LME, linear mixed-effects; PET, positron
emission tomography; ROI, region of interest; SUVR, standardized uptake valueratio.
4DISCUSSION
In the present study, we systematically explored the differences in
plasma NT1-tau levels throughout the clinical diagnosis and biological
stages of AD and how it related to Aβ, tau, neurodegeneration, and cog-
nitive decline using two independent cohorts, including cross-sectional
and longitudinal data of participants. First, we found that plasma NT1-
tau levels were specifically increased in symptomatic AD and showed
prominent abnormality during the late biological stage of the disease.
Plasma NT1-tau was positively associated with cerebral depositions of
Aβplaques and tau tangles in an Aβ-dependent manner. Meanwhile,
higher baseline NT1-tau could predict future greater tau aggregation
and faster cerebral Aβdeposition over time. Cross-sectionally, plasma
NT1-tau displayed an association with brain-wide regional GMV and
cortical thickness markedly superior to plasma BD-tau and NfL in the
AD trajectory and significantly correlated with AD-typical neurode-
generation and cognition in an Aβ-dependent manner. Finally, higher
baseline NT1-tau and BD-tau predicted disease progression as mea-
sured with more rapid AD-signature cortical thinning, hippocampal
atrophy, and cognitive decline. Altogether our findings suggest that
plasma NT1-tau is an Aβ-associated biomarker closely related to AD
pathophysiology, serving as a candidate predictor of downstream neu-
rodegeneration and cognitive decline. Our findings have important
implications for the utility of plasma NT1-tau in patient management
and monitoring in clinical practices.
Previous results with biofluid NT1-tau have already reported appar-
ent increases in symptomatic AD.21,24 Chen et al., who developed the
NT1-tau assay, preliminarily found that CSF and plasma NT1-tau lev-
els were elevated in MCI and dementia patients due to AD compared
with CSF Aβnegative CU cases in a small pilot clinical cohort.21 These
findings were confirmed in another observational study with a novel
NTB-tau assay to detect the same N-terminal tau fragments as NT1-
tau assay, showing that its levels in CSF were significantly higher in AD
cases but not in non-AD dementia.18 In this study, we further corrob-
orated previous findings by highlighting the specific plasma NT1-tau
elevations in symptomatic AD using two large, well-characterized
cohorts from China and the United States, indicating that plasma NT1-
tau increases may be attributed to AD pathophysiological alterations

10 of 14 LAN ET AL.
FIGURE 4 Cross-sectional association of plasma NT1-tau, BD-tau, and NfL with regional GMV and cortical thickness. Correlation coefficients
of plasma NT1-tau, BD-tau, and NfL with baseline regional GMV and cortical thickness across 68 brain regions in the whole cohort (A) and Aβ+
participants (B). The threshold was set at p<0.05 after Benjamini–Hochberg corrections. Aβ, amyloid beta; BD-tau, brain-derived tau; GMV, gray
matter volume; NfL, neurofilament light.
at the intermediate to later stages. This was supported by the A/T clas-
sification, where plasma NT1-tau was increased only in patients with
evidence of Aβand early tau pathologies.
In addition, we found that plasma NT1-tau was not increased in the
Aβ– CI group, which was also observed in another cohort study with a
small sample size, showing that plasma NT1-tau was specifically higher
in AD cases but not in non-AD dementia.24 However, this AD-specific
elevation should be interpreted cautiously. A cross-sectional observa-
tion study reported that NT1-tau in CSF and plasma were higher in
Creutzfeldt-Jakob disease than in AD.44 Consistently, another cohort
study also found that the level of this specific NT1-tau fragment in
CSF had more prominent elevations in Creutzfeldt-Jakob disease and
acute neurological disorders when compared to AD.18 Thus, NT1-tau
elevations could probably reflect intense neuronal injury and neurode-
generation in acute neurological disorders. Future investigations are
needed to elucidate the dynamic changes in NT1-tau during acute
neurological damage.
Advances in blood-based tests for tau biomarkers have concen-
trated predominantly on p-tau species (i.e., p-tau181, p-tau217, and
p-tau231), that become already abnormalities in preclinical AD align
with markedly tight associations with Aβplaques.45,46 Therefore,
plasma p-tau biomarkers significantly contribute to the early detection
and diagnosis of AD but limit their utility as biomarkers of insoluble tau
aggregation.5In contrast to p-tau, the levels of non-phosphorylated tau
species in biofluid seem to reach the threshold of abnormalities at the
later disease stages.10,11,18,23 For instance, MTBR-tau243 in CSF was
reported to be increased in symptomatic AD as a biomarker of AD-type
neocortical tau tangles that had the best correlation associated with
cognitive decline.10,47 Unlike p-tau biomarkers, MTBR-tau243 eleva-
tions across the AD continuum were better explained by tau pathology

LAN ET AL.11 of 14
FIGURE 5 Association of plasma NT1-tau, BD-tau, and NfL with longitudinal cortical thinning and hippocampal atrophy. Association of plasma
NT1-tau, BD-tau, and NfL with longitudinal changes in temporal-MetaROI cortical thickness (A) and residual HCV (B) in the GHABS cohort
(n=106). LME models were used with temporal-MetaROI cortical thickness, residual HCV as outcome, and the time ×plasma biomarkers
interaction as predictor with random intercepts and time slopes. Age and sex were used as covariates. For illustration purposes only, the regression
lines of longitudinal changes are shown for participants with plasma biomarker levels below versus above the median. ***p<0.001; **p<0.01;
*p<0.05. BD-tau, brain-derived tau; HCV, hippocampal volume; LME, linear mixed-effects; NfL, neurofilament light.
than by Aβpathology as measured with PET imaging.10 Thus, the dif-
ferential truncation patterns of tau fragments may provide different
aspects of AD-related pathological changes. The NT1-tau observation
in this study showed an Aβ-dependent correlation with the deposition
of Aβplaques and tau tangles. Notably, the associations were relatively
lower in the ADNI cohort, which may be explained by several factors.
Plasma NT1-tau levels could be saturated in the Aβ+participants of the
ADNI cohort (Figure 1),resultinginanattenuatedreactiontoadvanced
Aβand tau pathologies. In addition, the ADNI participants differed sig-
nificantly from the GHABS participants in terms of ethnicity, age, and
the prevalence of female and CI individuals, all of which may influence
plasma NT1-tau trajectory during the AD continuum. However, more
independent studies with longitudinal data are required to validate
these findings. Although we did not observe persistent NT1-tau asso-
ciation in Aβ+cases of the ADNI cohort, plasma NT1-tau was found
to be a significant predictor of more-rapid Aβaccumulation at follow-
up even with adjusting for baseline AβPET, particularly in the phase
of AD trajectory. This was consistent with previous results revealing
that NT1-tau fragments were actively generated and secreted along
with Aβplaque burden.48 Furthermore, we found a predictive effect
of plasma NT1-tau on greater tau aggregation in the future. This was
in line with a previous work by Chhatwal et al., showing that higher
plasma NT1-tau was related to faster longitudinal tau accumulation
in CU elderly of the HABS cohort.25 Together with the previous liter-
ature, our findings suggest that plasma NT1-tau is an Aβ-associated
biomarker for downstream AD-type pathological changes.
Plasma NT1-tau elevations were related to an increased risk of
developing AD dementia.24 Previous studies have reported the associ-
ations of Aβpathophysiology with plasma NT1-tau and BD-tau but not
with plasma NfL.23,25,49 Although both NT1-tau and BD-tau in plasma
showed better performance in predicting neurodegeneration than
plasma NfL in the AD continuum,23,25 no study has compared the per-
formance of these potential biomarkers in the same set of populations.
Here, we expanded previous findings by demonstrating that plasma
NT1-tau but not BD-tau showed significant correlations with baseline
GMV and cortical thickness across the whole brain regions, indicating
that plasma NT1-tau may reflect cross-sectional neurodegeneration
superior to that of plasma BD-tau. It was further confirmed by using
the markers of AD-typical neurodegeneration. These findings raise the
probability of NT1-tau as an N biomarker within the AT(N) frame-
work to improve the stratification of disease stages in clinical trials,
given that NfL is not an AD-specific neurodegeneration biomarker.49,50

12 of 14 LAN ET AL.
This hypothesis is supported by our findings that plasma NT1-tau is
a stronger predictor of longitudinal neurodegeneration and cognitive
decline than plasma NfL in the AD continuum, which is in line with pre-
vious work.25 It should be noted that the results of this pattern must be
interpreted carefully in the light of the study population. The GHABS
is a community-based cohort that focuses on Chinese Han older
adults within the Guangdong-Hong Kong-Macao Greater-Bay-Area of
South China26; thus the included participants have notably ethnic-
specific and area-specific characteristics. Further studies exploring
NT1-tau should indicate how it is affected by different aspects of the
AD pathology and whether it combined with p-tau biomarkers could
improve stratification and predict disease progression during the AD
trajectory.
The present study demonstrated the relationship of plasma NT1-
tau with AD pathophysiological features. The highlighted strengths of
this study are the systematic explorations of plasma NT1-tau across
the whole AD continuum using two independent populations with dif-
ferent social and ethnic backgrounds, and all participants are well
characterized with various biofluid marker measures, imaging scans,
and cognitive tests. A subset of samples had available longitudinal data
on MRI imaging and cognition, although the follow-up duration in the
GHABS cohort was slightly short. We would further validate these
findings using longitudinal data with more visits and longer follow-
up periods in the future. In addition, the lack of longitudinal AβPET
and tau PET imaging in the GHABS cohort did not allow us to con-
firm the predictive effect of plasma NT1-tau on subsequent Aβand
tau accumulations. Finally, participants in two cohorts had no available
CSF NT1-tau data, restricting the comparison between plasma and CSF
biomarkers on the AD trajectory.
In conclusion, our findings suggest that plasma NT1-tau is an AD-
specific biomarker that can predict subsequent disease progression
as measured with greater tau aggregation and faster Aβaccumula-
tion, brain atrophy, and cognitive decline. Moreover, plasma NT1-tau
appears to outperform plasma BD-tau and NfL in detecting Aβ-
associated neurodegeneration, highlighting the probability of plasma
NT1-tau as an AD-specific neurodegeneration biomarker for disease
stage classification and patient screening and monitoring in clinical tri-
als. This study provides novel insights into the role of plasma NT1-tau
in AD pathophysiology.
ACKNOWLEDGMENTS
The authors are grateful to Beth Ostaszewski and Elizabeth Head for
their assistance in establishing the plasma NT1-tau assay in our lab.
The authors would like to thank all the participants and staff of the
Greater-Bay-Area Healthy Aging Brain Study (GHABS) research group
(clinical trials No. NCT06183658) for their immense contributions to
data collection. The Shenzhen Bay Laboratory supercomputing center
supported the imaging processing. The authors would like to thank all
the Alzheimer’s Disease Neuroimaging Initiative (ADNI) participants
and staff for their contributions to data acquisition. The data collection
and sharing for this project were funded by the ADNI (National Insti-
tutes of Health Grant U01 AG024904) and DOD ADNI (Department
of Defense award number W81XWH-12-2-0012). ADNI is funded by
the National Institute on Aging, the National Institute of Biomed-
ical Imaging and Bioengineering, and through generous contribu-
tions from the following: AbbVie, Alzheimer’s Association; Alzheimer’s
Drug Discovery Foundation; Araclon Biotech; BioClinica, Inc.; Biogen;
Bristol-Myers Squibb Company; CereSpir, Inc.; Eisai Inc.; Elan Phar-
maceuticals, Inc.; Eli Lilly and Company; EuroImmun; F. Hoffmann-La
Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE
Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research
& Development, LLC.; Johnson & Johnson Pharmaceutical Research
& Development LLC.; Lumosity; Lundbeck; Merck & Co., Inc.; Meso
Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack Technologies;
Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging;
Servier; Takeda Pharmaceutical Company; and Transition Therapeu-
tics. The Canadian Institutes of Health Research is providing funds
to support ADNI clinical sites in Canada. Private sector contributions
are facilitated by the Foundation for the National Institutes of Health
(www.fnih.org). The grantee organization is the Northern California
Institute for Research and Education, and the study is coordinated
by the Alzheimer’s Disease Cooperative Study at the University of
California, San Diego. ADNI data are disseminated by the Labora-
tory for Neuro Imaging at the University of Southern California. This
study was funded by the National Natural Science Foundation of China
(Grant No. 82171197, 82301380), the Guangdong Basic and Applied
Basic Science Foundation for Distinguished Young Scholars (Grant
No.2023B1515020113), the Shenzhen Science and Technology Pro-
gram (Grant No. RCYX20221008092935096), National Key Research
and Development Program of China (2023YFC3605400), Shenzhen
Bay Laboratory (Grant No. S241101004-1), and Lingang Laboratory
(Grant No. LG-GG-202401-ADA070600).
CONFLICT OF INTEREST STATEMENT
The authors report no competing interests. Author disclosures are
available in the Supporting Information.
DATA AVAILABILITY STATEMENT
The data used in the current study were obtained from the GHABS
and ADNI cohorts. Derived data is available from the corresponding
author on request by any qualified investigator subject to a data use
agreement.
CONSENT STATEMENT
The GHABS and ADNI study was approved by institutional review
boards of all participating institutions, and written informed consent
was obtained for ethical considerations.
ORCID
Teng fe i Guo https://orcid.org/0000-0003-2982-0865
REFERENCES
1. Guo T, Dukart J, Brendel M, Rominger A, Grimmer T, Yakushev I.
Rate of β-amyloid accumulation varies with baseline amyloid bur-
den: implications for anti-amyloid drug trials. Alzheimer’s Dement.
2018;14:1387-1396. doi:10.1016/j.jalz.2018.05.013

LAN ET AL.13 of 14
2. Guo T, Shaw LM, Trojanowski JQ, Jagust WJ, Landau SM. Associa-
tion of CSF Aβ, amyloid PET, and cognition in cognitively unimpaired
elderly adults. Neurology. 2020;95:e2075-e2085. doi:10.1212/WNL.
0000000000010596
3. Cai Y, Du J, Li A, et al. Initial levels of β-amyloid and tau deposition
have distinct effects on longitudinal tau accumulation in Alzheimer’s
disease. Alzheimers Res Ther. 2023;15:1-14. doi:10.1186/s13195-023-
01178-w
4. Guo T, Korman D, La Joie R, et al. Normalization of CSF pTau measure-
ment by Aβ40 improves its performance as a biomarker of Alzheimer’s
disease. Alzheimers Res Ther. 2020;12:97. doi:10.1186/s13195-020-
00665-8
5. Jack CR Jr, Andrews JS, Beach TG, et al. Revised criteria for diag-
nosis and staging of Alzheimer’s disease: Alzheimer’s Association
Workgroup. Alzheimers Dement. 2024:20:5143-5169. doi:10.1002/alz.
13859
6. Mattsson-Carlgren N, Andersson E, Janelidze S, et al. Aβdeposi-
tion is associated with increases in soluble and phosphorylated tau
that precede a positive Tau PET in Alzheimer’s disease. Sci Adv.
2020;6:eaaz2387. doi:10.1126/sciadv.aaz2387
7. Therriault J, Vermeiren M, Servaes S, et al. Association of phospho-
rylated tau biomarkers with amyloid positron emission tomography
vs tau positron emission tomography. JAMA Neurol. 2023;80:188-199.
doi:10.1001/jamaneurol.2022.4485
8. Barthélemy NR, Li Y, Joseph-Mathurin N, et al. A soluble phosphory-
lated tau signature links tau, amyloid and the evolution of stages of
dominantly inherited Alzheimer’s disease. Nat Med. 2020;26:398-407.
doi:10.1038/s41591-020-0781-z
9. Karikari TK, Ashton NJ, Brinkmalm G, et al. Blood phospho-tau
in Alzheimer disease: analysis, interpretation, and clinical utility.
Nat Rev Neurol. 2022;18:400-418. doi:10.1038/s41582-022-00665-
2
10. Horie K, Salvadó G, Barthélemy NR, et al. CSF MTBR-tau243 is a spe-
cific biomarker of tau tangle pathology in Alzheimer’s disease. Nat Med.
2023;29:1954-1963. doi:10.1038/s41591-023-02443-z
11. Salvadó G, Horie K, Barthélemy NR, et al. Disease staging of
Alzheimer’s disease using a CSF-based biomarker model. Nat Aging.
2024;4:694-708. doi:10.1038/s43587-024-00599-y
12. Ashton NJ, Janelidze S, Al Khleifat A, et al. A multicentre valida-
tion study of the diagnostic value of plasma neurofilament light. Nat
Commun. 2021;12:3400. doi:10.1038/s41467-021- 23620-z
13. Hansson O. Biomarkers for neurodegenerative diseases. Nat Med.
2021;27:954-963. doi:10.1038/s41591-021-01382-x
14. Dage JL, Wennberg AMV, Airey DC, et al. Levels of tau pro-
tein in plasma are associated with neurodegeneration and cognitive
function in a population-based elderly cohort. Alzheimers Dement.
2016;12:1226-1234. doi:10.1016/j.jalz.2016.06.001
15. Mattsson N, Zetterberg H, Janelidze S, et al. Plasma tau in
Alzheimer disease. Neurology. 2016;87:1827-1835. doi:10.1212/
WNL.0000000000003246
16. Guo T, Li A, Sun P, et al. Astrocyte reactivity is associated with tau tan-
gle load and cortical thinning in Alzheimer’s disease. Mol Neurodegener.
2024;19:58. doi:10.1186/s13024-024-00750-8
17. Schindler SE, Galasko D, Pereira AC, et al. Acceptable performance of
blood biomarker tests of amyloid pathology—recommendations from
the Global CEO Initiative on Alzheimer’s Disease. Nat Rev Neurol.
2024;20:426-439. doi:10.1038/s41582-024-00977-5
18. Snellman A, Lantero-Rodriguez J, Emeršiˇ
c A, et al. N-terminal and
mid-region tau fragments as fluid biomarkers in neurological diseases.
Brain. 2022;145:2834-2848. doi:10.1093/brain/awab481
19. Gonzalez-Ortiz F, Turton M, Kac PR, et al. Brain-derived tau: a novel
blood-based biomarker for Alzheimer’s disease-type neurodegenera-
tion. Brain. 2023;146:1152-1165. doi:10.1093/brain/awac407
20. Schlegel K, Awwad K, Heym RG, et al. N368-Tau fragments gener-
ated by legumain are detected only in trace amount in the insoluble
Tau aggregates isolated from AD brain. Acta Neuropathol Commun.
2019;7:1-15. doi:10.1186/s40478-019-0831-2
21. Chen Z, Mengel D, Keshavan A, et al. Learnings about the com-
plexity of extracellular tau aid development of a blood-based screen
for Alzheimer’s disease. Alzheimer’s Dement. 2019;15:487-496. doi:10.
1016/j.jalz.2018.09.010
22. Rodriguez L, Rodriguez JL, Salvadó G, et al. Plasma N-terminal con-
taining tau fragments (NTA-tau): a biomarker of tau deposition in
Alzheimer’s disease. Mol Neurodegener. 2024;19:1-22. doi:10.1186/
s13024-024-00707-x
23. Gonzalez-Ortiz F, Kirsebom BE, Contador J, et al. Plasma brain-
derived tau is an amyloid-associated neurodegeneration biomarker
in Alzheimer’s disease. Nat Commun. 2024;15:1-13. doi:10.1038/
s41467-024-47286-5
24. Mengel D, Janelidze S, Glynn RJ, Liu W, Hansson O, Walsh DM. Plasma
NT1 tau is a specific and early marker of Alzheimer’s disease. Ann
Neurol. 2020;88:878-892. doi:10.1002/ana.25885
25. Chhatwal JP, Schultz AP, Dang Y, et al. Plasma N-terminal tau frag-
ment levels predict future cognitive decline and neurodegeneration in
healthy elderly individuals. Nat Commun. 2020;11:6024. doi:10.1038/
s41467-020-19543-w
26. Liu Z, Shi D, Cai Y, et al. Pathophysiology characterization of
Alzheimer’s disease in South China’s aging population: for the Greater-
Bay-Area Healthy Aging Brain Study (GHABS). Alzheimers Res Ther.
2024;16:84. doi:10.1186/s13195-024-01458-z
27. Stern AM, Van Pelt KL, Liu L, et al. Plasma NT1-tau and Aβ42 correlate
with age and cognitive function in two large Down syndrome cohorts.
Alzheimers Dement. 2023;19:5755-5764. doi:10.1002/alz.13382
28. Ashton NJ, Pascoal TA, Karikari TK, et al. Plasma p-tau231: a new
biomarker for incipient Alzheimer’s disease pathology. Acta Neu-
ropathol. 2021;141:709-724. doi:10.1007/s00401-021- 02275-6
29. Karikari TK, Emeršiˇ
c A, Vrillon A, et al. Head-to-head comparison of
clinical performance of CSF phospho-tau T181 and T217 biomarkers
for Alzheimer’s disease diagnosis. Alzheimers Dement. 2021;17:755-
767. doi:10.1002/alz.12236
30. Suárez-Calvet M, Capell A, Araque Caballero MÁ, et al. CSF progran-
ulin increases in the course of Alzheimer’s disease and is associated
with sTREM2, neurodegeneration and cognitive decline. EMBO Mol
Med. 2018;10:1-21. doi:10.15252/emmm.201809712
31. Lan G, Chen X, Yang J, et al. Microglial reactivity correlates to
presynaptic loss independent of β-amyloid and tau. Ann Neurol.
2024;95:917-928. doi:10.1002/ana.26885
32. Suárez-Calvet M, Morenas-Rodríguez E, Kleinberger G, et al. Early
increase of CSF sTREM2 in Alzheimer’s disease is associated with
tau related-neurodegeneration but not with amyloid-βpathology. Mol
Neurodegener. 2019;14:1-14. doi:10.1186/s13024-018-0301-5
33. Franzmeier N, Dehsarvi A, Steward A, et al. Elevated CSF GAP-
43 is associated with accelerated tau accumulation and spread in
Alzheimer’s disease. Nat Commun. 2024;15:202. doi:10.1038/s41467-
023-44374-w
34. Lan G, Cai Y, Li A, Liu Z, Ma S, Guo T. Association of presynap-
tic loss with Alzheimer’s disease and cognitive decline. Ann Neurol.
2022;92:1001-1015. doi:10.1002/ana.26492
35. Desikan RS, Ségonne F, Fischl B, et al. An automated labeling system
for subdividing the human cerebral cortex on MRI scans into gyral
based regions of interest. Neuroimage. 2006;31:968-980. doi:10.1016/
j.neuroimage.2006.01.021
36. Lan G, Li A, Liu Z, Ma S, Guo T. Presynaptic membrane protein dys-
function occurs prior to neurodegeneration and predicts faster cogni-
tive decline. Alzheimers Dement. 2023;19:2408-2419. doi:10.1002/alz.
12890
37. Guo T, Korman D, Baker SL, Landau SM, Jagust WJ. Longitudinal cog-
nitive and biomarker measurements support a unidirectional pathway
in Alzheimer’s disease pathophysiology. Biol Psychiatry. 2021;89:786-
794. doi:10.1016/j.biopsych.2020.06.029

14 of 14 LAN ET AL.
38. Landau SM, Fero A, Baker SL, et al. Measurement of longitudinal β-
amyloid change with 18 F-Florbetapir PET and standardized uptake
value ratios. JNuclMed. 2015;56:567-574. doi:10.2967/jnumed.114.
148981
39. Li A, Zhao R, Zhang M, et al. [18F]-D3FSP β-amyloid PET imaging
in older adults and Alzheimer’s disease. Eur J Nucl Med Mol Imaging.
2024;51:3990-4000. doi:10.1007/s00259-024-06835-2
40. Du J, Li A, Shi D, et al. Association of APOE-ε4, osteoarthritis, β-
Amyloid, and tau accumulation in primary motor and somatosensory
regions in Alzheimer disease. Neurology. 2023;101:e40-e49. doi:10.
1212/WNL.0000000000207369
41. Maass A, Landau S, Baker SL, et al. Comparison of multiple tau-PET
measures as biomarkers in aging and Alzheimer’s disease. Neuroimage.
2017;157:448-463. doi:10.1016/j.neuroimage.2017.05.058
42. Crane PK, Carle A, Gibbons LE, et al. Development and assessment of a
composite score for memory in the Alzheimer’s Disease Neuroimaging
Initiative (ADNI). Brain Imaging Behav. 2012;6:502-516. doi:10.1007/
s11682-012-9186-z
43. Gibbons LE, Carle AC, Mackin RS. A composite score for executive
functioning, validated in Alzheimer’s Disease Neuroimaging Initiative
(ADNI) participants with baseline mild cognitive impairment. Brain
Imaging Behav. 2012;6:517-527. doi:10.1007/s11682-012- 9176-1
44. Mengel D, Mok TH, Nihat A, et al. Nt1-tau is increased in CSF and
plasma of CJD patients, and correlates with disease progression. Cells.
2021;10:1-13. doi:10.3390/cells10123514
45. Karikari TK, Pascoal TA, Ashton NJ, et al. Blood phosphorylated tau
181 as a biomarker for Alzheimer’s disease: a diagnostic perfor-
mance and prediction modelling study using data from four prospec-
tive cohorts. Lancet Neurol. 2020;19:422-433. doi:10.1016/S1474-
4422(20)30071-5
46. Ashton NJ, Janelidze S, Mattsson-Carlgren N, et al. Differential roles
of Aβ42/40, p-tau231 and p-tau217 for Alzheimer’s trial selection
and disease monitoring. Nat Med. 2022;28:2555-2562. doi:10.1038/
s41591-022-02074-w
47. Nelson PT, Alafuzoff I, Bigio EH, et al. Correlation of Alzheimer disease
neuropathologic changes with cognitive status: a review of the liter-
ature. J Neuropathol Exp Neurol. 2012;71:362-381. doi:10.1097/NEN.
0b013e31825018f7
48. Sato C, Barthélemy NR, Mawuenyega KG, et al. Tau kinetics in neu-
rons and the human central nervous system. Neuron. 2018;97:1284-
1298.e7. doi:10.1016/j.neuron.2018.02.015
49. Mattsson N, Insel PS, Palmqvist S, Portelius E, Zetterberg H. Cere-
brospinal fluid tau, neurogranin, and neurofilament light in Alzheimer’s
disease. EMBO Mol Med. 2016;8:1184-1196. doi:10.15252/emmm.
201606540
50. Benussi A, Karikari TK, Ashton N, et al. Diagnostic and prognostic
value of serum NfL and p-Tau(181) in frontotemporal lobar degener-
ation. J Neurol Neurosurg Psychiatry. 2020;91:960-967. doi:10.1136/
jnnp-2020-323487
SUPPORTING INFORMATION
Additional supporting information can be found online in the Support-
ing Information section at the end of this article.
How to cite this article: Lan G, Zhang L, Li A, et al. Plasma
N-terminal tau fragment is an amyloid-dependent biomarker in
Alzheimer’s disease. Alzheimer’s Dement. 2025;21:e14550.
https://doi.org/10.1002/alz.14550