https://doi.org/10.

1093/braincomms/fcaf221 BRAIN COMMUNICATIONS 2025: fcaf221 | 1

BRAIN COMMUNICATIONS
REPORT
The potential dual role of tau phosphorylation:
plasma phosphorylated-tau217 in newborns

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

and Alzheimer’s disease
Fernando Gonzalez-Ortiz,1,2,3 Jakub Vávra,1 Emma Payne,4 Bjørn-Eivind Kirsebom,5,6,7
Ulrika Sjöbom,8,9 Cristiano Santos,2 Jordi Júlvez,10 Kaitlin Kramer,11,12 David Zalcberg,11,12
Laia Montoliu-Gaya,1 Michael Turton,13 Peter Harrison,13 Ann Hellström,9
Henrik Zetterberg,1,2,14,15,16,17 Tormod Fladby,7,18 Marc Suárez-Calvet,19,20,21
Robert D. Sanders11,12,22,23 and Kaj Blennow1,2,24,25

Tau phosphorylation plays an important role in brain physiology and pathology. During foetal development, it supports microtubule
dynamics and neuroplasticity, whereas in Alzheimer’s disease (AD), it drives pathological tau aggregation and tangle formation. In
this multicentre study (n = 462), we measured plasma phosphorylated-tau217 in healthy newborns, premature infants, patients
with AD and healthy controls across various age groups. Plasma phosphorylated-tau217 levels were significantly higher in newborns
compared to healthy individuals of any age group and even exceeded levels observed in patients with AD. In newborns, plasma
phosphorylated-tau217 levels inversely correlated with perinatal factors such as gestational age. Longitudinal analysis of preterm
infants demonstrated a decline in serum phosphorylated-tau217 levels over the first months of life, approaching levels observed in
young adults. In contrast, elevated plasma phosphorylated-tau217 in older individuals was associated with AD pathology. Our
findings corroborate the crucial role of tau phosphorylation in early brain development. However, in AD, tau phosphorylation
transitions into a pathological mechanism. The high levels of blood-based phosphorylated-tau217 observed at birth and subsequent
clearance might indicate distinct regulatory mechanisms that prevent tau aggregation in early life. Further studies are needed to
explore the shared mechanisms of tau phosphorylation in newborns and AD.

1 Department of Psychiatry and Neurochemistry, Institute of Neuroscience and Physiology, The Sahlgrenska Academy at the
University of Gothenburg, Gothenburg 417 11, Sweden
2 Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölndal 431 80, Sweden
3 Neurology Department, Neurocode USA Inc, Bellingham, WA 98225, USA
4 St George Hospital, South Eastern Sydney Local Health District, Sydney, New South Wales 2217, Australia
5 Department of Neurology, University Hospital of North Norway, Tromsø 9038, Norway
6 Department of Psychology, Faculty of Health Sciences, The Arctic University of Norway, Tromsø 9037, Norway
7 Department of Neurology, Akershus University Hospital, Lørenskog 1478, Norway
8 Learning and Leadership for Health Care Professionals at the Institute of Health and Care Science at Sahlgrenska Academy at
University of Gothenburg, Gothenburg 417 11, Sweden
9 Department of Clinical Neuroscience at the Institution of Neuroscience and Physiology at Sahlgrenska Academy at University of
Gothenburg, Gothenburg 417 11, Sweden
10 Clinical and Epidemiological Neuroscience Group (NeuroÈpia), Institut d’Investigació Sanitària Pere Virgili (IISPV), Reus 43204,
Spain
11 Central Clinical School, Faculty of Medicine and Health, The University of Sydney, Sydney, New South Wales 2006, Australia
12 Department of Anaesthetics, Royal Prince Alfred Hospital, Sydney Local Health District, Sydney, New South Wales 2050,
Australia

Received February 27, 2025. Revised May 19, 2025. Accepted June 06, 2025. Advance access publication June 7, 2025
© The Author(s) 2025. Published by Oxford University Press on behalf of the Guarantors of Brain.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse,
distribution, and reproduction in any medium, provided the original work is properly cited.
2 | BRAIN COMMUNICATIONS 2025, fcaf221 F. Gonzalez-Ortiz et al.

13 Bioventix Plc, Farnham, Surrey GU9 7SX, UK

14 Department of Neurodegenerative Disease, UCL Institute of Neurology, London WC1N 3BG, UK
15 UK Dementia Research Institute at UCL, London WC1N 3BG, UK
16 Hong Kong Center for Neurodegenerative Diseases, Hong Kong 200020, China
17 Wisconsin Alzheimer’s Disease Research Center, University of Wisconsin School of Medicine and Public Health, University of
Wisconsin-Madison, Madison, WI 53 792, USA
18 University of Oslo, Institute for Clinical Medicine, Campus Ahus, Oslo 1478, Norway
19 Barcelonaβeta Brain Research Center (BBRC), Pasqual Maragall Foundation, Barcelona 08005, Spain
20 Hospital del Mar Research Institute, Barcelona 08003, Spain
21 Servei de Neurologia, Hospital del Mar, Barcelona 08003, Spain
22 NHMRC Clinical Trials Centre, The University of Sydney, Sydney, New South Wales 2006, Australia
23 Institute of Academic Surgery, Royal Prince Alfred Hospital, Sydney Local Health District, Sydney, New South Wales 2050,

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

Australia
24 Paris Brain Institute, ICM, Pitié-Salpêtrière Hospital, Sorbonne University, Paris 75013, France
25 Neurodegenerative Disorder Research Center, Division of Life Sciences and Medicine, and Department of Neurology, Institute on
Aging and Brain Disorders, University of Science and Technology of China and First Affiliated Hospital of USTC, Hefei 230026,
China
Correspondence to: Fernando Gonzalez-Ortiz
Clinical Neurochemistry Lab House V3, Mölndal Hospital
Floor 2, Biskopsbogatan 27
SE-43180 Mölndal, Sweden
E-mail: [email protected]
Keywords: phosphorylated tau; newborns; plasma biomarkers; Alzheimer’s disease
Plasma p-tau217 is higher in newborns than in AD BRAIN COMMUNICATIONS 2025, fcaf221 | 3

Graphical Abstract

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

Introduction specific sites.1,2 In the foetal brain, tau is predominantly ex­

pressed as a shorter isoform lacking exon 10, which results in
Tau is a critical protein involved in regulating microtubule a form with three microtubule-binding repeats (3R tau).3 In
dynamics, axonal transport and neurite outgrowth, with contrast, the adult brain expresses a mixture of both 3R and
each function being modulated by phosphorylation at 4R tau isoforms, the latter including exon 10, which adds a
4 | BRAIN COMMUNICATIONS 2025, fcaf221 F. Gonzalez-Ortiz et al.

fourth microtubule-binding region.3 In addition to differ­

ences in microtubule-binding repeats, foetal tau predomin­
Materials and methods
antly contains zero N-terminal inserts (0N), while adult Cohort-1 (ALFA age): The use of umbilical cord blood sam­
tau includes all six isoforms with varying numbers of ples was approved by the Ethics Committee (CEIm) of Vall
N-terminal repeats (1N, or 2N), contributing to its function­ d’Hebron University Hospital on 27 July 2018, and by the
al diversity.3,4 scientific committee and the scientific management of the
Dysregulations in tau phosphorylation are strongly impli­ Banc de Sang i Teixits (BST) on 16 July 2018. This study
cated in tau dysfunction,1,5 a key feature in Alzheimer’s dis­ was approved by the Independent Ethics Committee ‘Parc
ease, where abnormal phosphorylation contributes de Salut Mar,’ Barcelona (2015/6026/I). The ALFA Age
significantly to disease progression.6 In Alzheimer’s disease, study (2018/8089/I) was approved by the Independent
tau phosphorylation not only leads to a loss of function Ethics Committee ‘Parc de Salut Mar,’ Barcelona.

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

but may also introduce harmful gain-of-function effects, Cohort-2: Ethics approval was obtained from SLHD
such as making tau prone for aggregation with the subse­ Research and Ethics Committee [Royal Prince Alfred
quent development of tangle pathology.7-9 In contrast, in Hospital (RPAH) zone]. Trained recruiters obtained written,
the developing foetal brain, high phosphorylated tau informed consent in English for all maternal participants in
(p-tau) levels correlate with dynamic microtubule activity antenatal visits or on the birthing unit. Participants were ex­
during key periods of brain plasticity and supports healthy cluded if they were non-English speaking, or had a history of
neuronal development without forming tangles.10,11 psychological illness or other conditions that may interfere
Moreover, increased p-tau concentrations in CSF from new­ with capacity to provide informed consent after appropriate
borns have been reported.12 However, it is unknown counselling in English. The study was performed according
whether the heightened phosphorylation of tau observed in to the National Statement on Ethical Conduct in Human
foetal brains and CSF from newborns can also be detected Research (2007) (link to national statement) and the
in peripheral blood using the currently available immunoas­ CPMP/ICH Note for Guidance on Good Clinical Practice
says and whether plasma p-tau levels can serve as clinically (link to CPMP/ICH); approval number: 2022/ETH01100.
relevant markers in newborns. Cohort-3 (DDI) has been approved by the Regional
Levels of plasma p-tau, particularly p-tau217, have Committees for Medical and Health Research Ethics in
been effectively used in blood-based assays to identify Norway (REK: 2013/150). All participants gave a written in­
Alzheimer’s disease pathology even in the disease’s preclinic­ formed consent before participating in the study.
al stages.13,14 Due to a clear increase in Alzheimer’s Cohort-4 (premature newborns) from a randomized con­
disease and strong association with amyloid markers, trolled trial was approved by the Regional Ethical Board in
especially amyloid PET positivity, the term ‘amyloid- Gothenburg (Dnr 303-11) and registered at ClinicalTrials.
dependent’ is often applied to describe blood-based p-tau gov (NCT02760472). Infants were included after oral and
markers.15,16 Plasma p-tau217 also correlates with increas­ written informed consent from parents or legal guardians.
ing severity of tau pathology measured by tau PET17
and tangle load at autopsy18 serving not only as a diagnostic
but also as a prognostic marker in Alzheimer’s disease.19
Participant characteristics
However, the dependency of p-tau to amyloid has Cohort-1: ALFA age
been debated extensively.20-22 For instance, we and other Cohort-1 is an observational and cross-sectional study con­
groups have shown that plasma p-tau can be an ducted at the Barcelonaβeta Brain Research Center
amyloid-independent marker in neurodegenerative disorders (BBRC), and its main aim was to assess the effect of aging
such as Niemann–Pick disease type C, Creutzfeldt–Jakob in blood biomarkers in different age groups. This study com­
disease and amyotrophic lateral sclerosis, and that a pared four age groups: (i) umbilical cord blood from new­
temporary increase in plasma p-tau is seen also in acute borns; (ii) plasma from teenagers (12–17 years old); (iii)
neurological conditions such as traumatic brain injury and plasma from young adults (18–25 years old); and (iv) plasma
cardiac arrest, with higher levels predicting poorer clinical from old adults (≥70). The human umbilical cord blood was
long-term outcomes.22-27 Altogether, these findings obtained from the Catalan blood and tissue bank. The inclu­
suggest that tau phosphorylation serves a more complex sion criteria of the study was the following: (i) males and fe­
role beyond its association with Alzheimer’s disease males between 18 and 25 years (young adults group) or ≥70
pathology.20 years for persons at the time of inclusion (old adults group);
Understanding the key physiological brain mechanisms (ii) subjects with no subjective cognitive complaints; (iii) in­
that protect the foetal brain from phosphorylated tau-related dividuals interested in participating in the study who fully
pathology may offer insights into therapeutic strategies to understand all the procedures that will be performed; (iv) ex­
prevent or reverse tau pathology in neurodegenerative dis­ plicit participant agreement to undergo all the study proce­
eases.4,11 Here, we aim to investigate phosphorylated tau dures, which encompass: collection of basic demographic
in newborns and Alzheimer’s disease by comparing levels data and collection of a blood sample; and (v) give informed
of blood-based p-tau217 in different age groups and consent and agree that no data resulting from the study
Alzheimer’s disease. (which is no clinically relevant) will be given to the
Plasma p-tau217 is higher in newborns than in AD BRAIN COMMUNICATIONS 2025, fcaf221 | 5

Table 1 Demographic information for cohorts 1–3

ALFA age cohort

Teenagers Young adults Older adults

(12) (28) (28)
Age, years NA 20.9 (1.9) 73.9 (1.9)
Mean (SD) [range] [19–26] [70–77]
Female, n (%) 6 (50.0) 10 (35.7) 13 (46.4)
Plasma p-tau217 pg/mL, mean (SD) 1.8 (0.4) 1.4 (0.7) 1.8 (1.3)

DDI cohort

Controls A− (56) MCI A+ (147) Dementia A+ (16)

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

Age, years 61.2 (9.8) 67.7 (8.0) 66.6 (8.9)
Mean (SD) [range] [41–80] [50–83] [51–83]
Female, n (%) 35 (62.5) 76 (51.7) 6 (37.5)
MMSE, median (IQR) 30 (1) 27 (3) 22 (4)
Plasma p-tau217 pg/mL, mean (SD) 1.7 (0.7) 3.6 (1.65) 4.6 (2.4)

Newborns

ALFA age cohort RPAH cohort

(55) (106)
Gestational age (weeks), mean (SD) 39.7 (0.8) 38.2 (2.2)
Female, n (%) 30 (54.5) NA
Plasma p-tau217 pg/mL, mean (SD) 10.2 (3.9) 9.2 (4.9)
A+/−, positive or negative CSF marker for amyloid plaques; N+/−, positive or negative marker for neurodegeneration; MCI, mild cognitive impairment; SD, standard deviation; IQR,
interquartile range; n, number of cases; %, percentage; DDI, Dementia Disease Initiation; RPAH, Royal Prince Alfred Hospital.

participant. The exclusion criteria were the following: (i) period, either on the birthing unit or in operating theatres.
cognitive impairment shown by a MMSE < 26; (ii) any sig­ A total of 1–3 mL samples were collected in addition to stand­
nificant systemic illness or unstable medical condition that ard samples and centrifuged in the RPAH Department of
could lead to difficulty complying with the protocol [includ­ Anaesthetics laboratory, and plasma samples were then stored
ing auditive and visual impairment, renal insufficiency under in deidentified cryovials. The inclusion and exclusion criteria
haemodialysis treatment, hepatic cirrhosis, chronic pneumo­ are described in an earlier study.29 Demographic characteris­
pathy under oxygen therapy, solid organ transplantation, tics of the participants can be found in Table 1.
fibromyalgia, active oncologic disease under treatment (ex­
cluding localized tumours)]; (iii) any significant major psy­ Cohort-3: dementia disease initiation
chiatric illness (following DSM-IV diagnosis manual) or Healthy controls and Alzheimer’s disease patients for the
diseases that interfere with cognitive function (including ma­ current project were sourced from the Norwegian
jor depression disorder, bipolar disorder, schizophrenia); (iv) Dementia Disease Initiation (DDI) cohort database. The
acquired brain injury: brain traumatic injury with parenchy­ Norwegian multicentre study DDI cohort comprise non-
mal or extra-axial macroscopic injury, large vessel ischaemic demented participants aged between 40 and 80 years with
stroke or haemorrhagic stroke, brain tumours or other con­ a native language of either Norwegian, Danish or Swedish.
ditions that may cause acquired brain injury (brain radio- or Participants were primarily recruited from memory clinics
chemotherapy); (v) Parkinson’s disease, epilepsy under treat­ and advertisements in local news media, recruited at univer­
ment and with frequent seizures (>1/month) in the last year, sity hospitals across Norway between 2013 and 2022.
multiple sclerosis or any other neurodegenerative disease; Demographic characteristics of the participants can be found
and (vi) researcher criteria: individuals that have any condi­ in Table 1. A detailed description of inclusion and exclusion
tion that, under researcher’s view, could lead to difficulty criteria is outlined in a previous publication.30
complying with the protocol. All participants in the ALFA
Age study signed the study’s informed consent form that Cohort-4: premature newborns
had also been approved by the Independent Ethics
Cohort-4 includes 14 out of 90 extremely preterm infants
Committee ‘Parc de Salut Mar,’ Barcelona.28 Demographic
(<28 weeks gestational age) from a randomized interven­
characteristics of the participants can be found in Table 1.
tional study at the Queen Silvia Children’s Hospital neonatal
intensive care unit between April 2013 and September 2015.
Cohort-2: RPAH Infants were randomized to different parenteral nutrition
In cohort-2, umbilical venous cord blood was collected by a strategies regarding lipids, either an omega-3-enriched lipid
trained attending midwife in the immediate postpartum emulsion (SMOF-lipid) or standard care (Clinoleic). Blood
6 | BRAIN COMMUNICATIONS 2025, fcaf221 F. Gonzalez-Ortiz et al.

samples were collected as serum from the umbilical cord and newborns (n = 14) with longitudinal measurements up to
at postnatal Days 1, 7, 14 and 28 and at postmenstrual ages Day 133 after birth. The analyses performed in this cohort
32, 36 and 40 weeks. The inclusion and exclusion in the aimed to investigate longitudinal changes in serum
study are described earlier.31 p-tau217 after birth in premature newborns. Demographic
characteristics for cohort-1–3 can be found in demographic
Table 1. Demographic information for cohort-4 can be
Sample collection and biomarker found in Supplementary Table 1. Additionally, Cohort 1 in­
measurements cluded an additional set of samples (n = 111) with biomarker
Plasma and serum samples were obtained according to data for plasma NfL, total tau, Aβ42 and Aβ40. However,
standard procedures and stored at −80°C until use. due to volume constraints, plasma p-tau217 levels were
Blood-based p-tau217 on the Simoa HD-X platform using not measured in this sample set, precluding direct compari­

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

the University of Gothenburg in-house protocol was previ­ sons between p-tau217 and other biomarkers. The demo­
ously described.13 In cohort-1, plasma neurofilament light graphic characteristics of these participants are presented
chain (NfL), total tau, amyloid42 and amyloid40 were mea­ in Supplementary Table 2.
sured using Quanterix commercial kits (103670) in a subset
of participants. Plasma brain-derived tau (BD-tau) in Plasma p-tau217 levels across age
cohort-2 was measured using the Gothenburg University in- groups
house method.32 Plasma p-tau217 measurements in cohorts
1–4 were performed between January and September 2024. In cohort-1, levels of plasma p-tau217 in newborns were sig­
Signal variations within and between analytical runs were as­ nificantly higher in the healthy newborns group (n = 55,
sessed using two internal quality control plasma samples at 10.19 ± 3.92 pg/mL) compared with teenagers (n = 12, 1.79
the beginning and the end of each run. ± 0.43 pg/mL), young adults (n = 28, 1.33 ± 0.69 pg/mL)
and older adults (n = 28, 1.78 ± 1.31 pg/mL). No significant
differences were observed in levels of plasma p-tau217 be­
Statistical analysis tween teenagers, young and old adults (Fig. 1, P > 0.99).
Statistical analyses were performed with Prism, version 9.3.1 Moreover, the results observed in newborns in cohort-1
(GraphPad) and RStudio (R version 4.3.2). The distributions were replicated in cohort-2 (n = 106). No significant differ­
of data sets were examined for normality using the ences were observed between levels of plasma p-tau217 in
Kolmogorov–Smirnov test. Because the data were nonnor­ newborns in cohort-1 and cohort-2 (10.19 ± 3.92 pg/mL ver­
mally distributed, nonparametric tests were used. P-values sus 9.14 ± 4.87 pg/mL, P > 0.99, Fig. 1).
(including those adjusted for multiple comparisons) were Cohort 1 included a set of participants with biomarker
considered significant at the two-sided P < 0.05 level. Fold data for plasma NfL, total tau, Aβ42 and Aβ40, but without
changes were examined by comparing biomarker values plasma p-tau217. Although direct comparisons with
with the means of the control groups. p-tau217 were not possible, we observed a similar pattern
between plasma total tau (T-tau) and p-tau217, with both
showing significantly higher levels in newborns compared
to other groups (P < 0.0001; Fig. 1 and Supplementary
Results Fig. 1). Conversely, plasma Aβ42 and Aβ40 levels were sig­
nificantly lower in the newborn group compared to the other
Cohort characteristics groups (P < 0.0001; Supplementary Fig. 1). Plasma NfL le­
‘Cohort-1’ was the ALFA age cohort (n = 123) that included vels were significantly higher in newborns than in teenagers
individuals in different age groups (newborns, n = 55; teen­ (P = 0.0002), yet lower than in older adults (P = 0.0027).
agers, n = 12; young adults, n = 28; older adults, n = 28).
Cross-sectional analyses in this cohort allowed us to investi­ Associations of cord blood p-tau217
gate the group differences in levels of plasma p-tau217 at dif­
ferent ages. The plasma p-tau217 results in cohort-1 were
with perinatal factors
independently validated in a clinical cohort (‘cohort-2’), Cohort-2 included extensive clinical data, enabling the ex­
which consisted of neonatal patients from the RPAH, ploration of associations between p-tau217 and perinatal
Sydney (n = 106). Moreover, plasma p-tau217 results in factors. These findings build upon a previous publication
cohort-1 and -2 were compared against the Dementia that examined the relationship between Alzheimer’s disease
Disease Initiation cohort (‘cohort-3’) that included plasma biomarkers and neonatal hypoxia, where we reported that
samples from individuals with CSF positive biomarkers for cord blood BD-tau is associated with indicators of foetal
Alzheimer’s disease with either mild cognitive impairment hypoxia, while p-tau217 levels may reflect a more general
or Alzheimer’s disease dementia (n = 163) and controls (n susceptibility to neurodevelopmental issues.29 In the present
= 56). Alzheimer’s disease was defined by CSF biomarker study, we contextualize these associations within biomarker
positivity (A+, as defined by CSF Aβ42/40 < 0.077). changes observed in older individuals and Alzheimer’s
Finally, ‘cohort-4’ included serum samples from premature disease.
Plasma p-tau217 is higher in newborns than in AD BRAIN COMMUNICATIONS 2025, fcaf221 | 7

compared p-tau217 levels in newborns from cohort-1

(n = 55) and cohort-2 (n = 106) with controls (n = 56) and
patients with Alzheimer’s disease (n = 163) from the
Dementia Disease Initiation cohort (cohort-3).
The control group exhibited the lowest levels of plasma
p-tau217 with concentration being similar to the ones in
the older adults group in cohort-1 (1.66 ± 0.71 pg/mL versus
1.78 ± 1.31 pg/mL). In comparison, plasma p-tau217 levels
in the Alzheimer’s disease group showed a significant in­
crease (3.68 ± 1.75 pg/mL) and a high accuracy at differenti­
ating Alzheimer’s disease from controls (Area under the

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

curve: 0.88, Supplementary Fig. 2). However, levels of
p-tau217 in newborns from cohort-1 (10.19 ± 3.92 pg/mL)
and cohort-2 (9.14 ± 4.87 pg/mL) were significantly higher
than levels in Alzheimer’s disease (3.68 ± 1.75 pg/mL). See
Fig. 2A.
When using the control group from cohort-3 as reference,
we observed the following fold changes, 2.8 for Alzheimer’s
Figure 1 Plasma p-tau217 levels across different age disease participants, 12 for newborns in cohort-1 and 10.54
groups. The figure shows a comparison of plasma p-tau217
for newborns in Cohort 2. These results are presented as
concentrations (pg/mL) in newborns from cohort-1 (n = 55),
Z-scores in Fig. 2B.
newborns from cohort-2 (n = 106), teenagers (n = 12), young
adults (n = 28) and older adults (n = 28). Group differences were
The lack of significant difference in p-tau217 levels be­
examined using the Mann–Whitney test (two categories) or the tween newborns in cohort-1 and cohort-2 (Fig. 2, P >
Kruskal–Wallis test with Dunn’s multiple comparisons (three 0.99), and between the older adults’ group in cohort-1 and
groups). In each box plot, the horizontal bar on top of the coloured the control group in cohort-3 (P > 0.99), indicates consistent
area shows the 75% percentile, the middle bar depicts the median, plasma p-tau217 levels across the different cohorts.
and the lower bar shows the 25% percentile. Values that are above
the 75% percentile and below the 25% percentile are shown
outside the coloured area. Each individual data point represents
Longitudinal changes of serum
plasma p-tau217 concentrations measured from a single participant p-tau217 in premature newborns
within the respective age group.
In cohort-4, we evaluated longitudinal changes in serum
p-tau217 levels in 14 extremely premature preterm new­
borns (<28 weeks gestational age) over time, measured in
In cohort-2, levels of p-tau217 in umbilical cord blood were postnatal weeks. In Fig. 3, each curve represents a single pa­
negatively associated with gestational age (Spearman’s tient, with dots marking p-tau217 levels at specific time
rho = −0.25, P = 0.0096), birth weight (Spearman’s rho = points. Despite this overarching pattern, there is notable
−0.23, P = 0.019) and head circumference (Spearman’s inter-patient variability in both the initial levels of
rho = −0.24, P = 0.016). However, only gestational age p-tau217 and the rate of decline. Some newborns exhibit
remained significant after multivariable analysis.29 Notably, markedly high levels shortly after birth, with steep reduc­
data on cord BD-tau and NfL revealed no significant associa­ tions, while others demonstrate lower starting levels and a
tions with perinatal factors.29 This distinction emphasizes the more gradual decrease. By the later stages of postnatal devel­
differences in dynamics between phosphorylated and non- opment (Days 100–140), p-tau217 levels in all patients ap­
phosphorylated forms of tau despite showing similar increases proach a stable, low range, suggesting the establishment of
in newborns. a baseline level as the brain matures.
Additionally, while the method of delivery significantly in­
fluenced BD-tau and NfL levels—caesarean section being as­
sociated with the highest increases in these markers—no
such variation was observed for p-tau217, which remained
Discussion
consistent across delivery methods.29 Tau hyperphosphorylation not always leads to aggregation
but rather seems to play a dual role in the brain.33
Phosphorylation sites like serine-262 have been widely stud­
Plasma p-tau217 levels in healthy ied in the context of brain development and more recently re­
controls, Alzheimer’s disease and cognized as a critical epitope involved in tau pathology and
tangle formation in Alzheimer’s disease,9,11 suggesting
newborns shared mechanisms in brain physiology and pathology.
To investigate how plasma p-tau217 concentrations com­ Here, we demonstrate that increased levels of phosphory­
pared between newborns and Alzheimer’s disease, we lated tau can also be detected in the peripheral blood of
8 | BRAIN COMMUNICATIONS 2025, fcaf221 F. Gonzalez-Ortiz et al.

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

Figure 2 Levels of plasma p-tau217 in healthy controls, Alzheimer’s disease and newborns. A shows a comparison of plasma
p-tau217 concentrations (pg/mL) among four groups: controls (n = 56), Alzheimer’s disease cohort (n = 163), newborns cohort-1 (n = 55) and
newborns cohort-2 (n = 106). Group differences were examined using the Mann–Whitney test (two categories) or the Kruskal–Wallis test with
Dunn’s multiple comparisons (three groups). In each box plot, the horizontal bar on top of the coloured area shows the 75% percentile, the middle
bar depicts the median, and the lower bar shows the 25% percentile. Values that are above the 75% percentile and below the 25% percentile are
shown outside the coloured area. Each individual data point represents the plasma p-tau217 concentration measured from a single participant
within the respective group. (B) Z-score-transformed plasma p-tau217 concentrations are shown for the control group (cohort-3), the
Alzheimer’s disease group (cohort-3) and newborns (cohort-1 and cohort-2). Group differences in this panel were also assessed using the
Kruskal–Wallis test with Dunn’s multiple comparisons test.

decline in blood-based p-tau levels during the first months of

life, eventually reaching levels similar to those in young adults.
Moreover, in two independent cohorts, levels of p-tau217 in
newborns were significantly higher than in Alzheimer’s dis­
ease. Our findings show that the increased levels of blood-
based p-tau217 in newborns can be indicative of its physio­
logical role in normal neurodevelopment.2,4,34 This is further
supported by the associations between umbilical cord blood
p-tau217 levels and perinatal variables such as gestational
age that underscore the relevance of tau phosphorylation in
foetal brain development.4 This feat was unique to phos­
phorylated tau in our data set since we observed no associa­
tions between perinatal variables and non-phosphorylated
tau markers, such as BD-tau, or NfL.29
In line with these results, longitudinal measurements of
serum p-tau217 in premature newborns revealed that this
marker remains elevated for several weeks after birth before
Figure 3 Longitudinal trajectories of serum p-tau217 in declining towards a ‘baseline level’ found in young adults.
preterm infants < 28 weeks gestational age at birth. Spaghetti This persistence aligns with the dynamic regulation of tau
plot showing the longitudinal changes in serum p-tau217 levels in phosphorylation during the early stages of postnatal brain
individual preterm infants over time (n = 14), measured in postnatal development, potentially driven by ongoing neuroplasticity
weeks. Each curve represents a single patient, with dots marking and the maturation of enzymatic pathways involved in tau
serum p-tau217 levels at specific time points. regulation.10,11 The variability in decline rates among indivi­
duals further points to the influence of inter-patient differ­
ences in developmental trajectories. These differences in
newborns and that plasma p-tau217 levels are significantly trajectories might reflect the dynamic regulation of tau phos­
higher in newborns compared to healthy individuals across phorylation during early development, likely driven by in­
different age groups. Importantly, longitudinal data show a creasing maturation of kinases, phosphatases and the
Plasma p-tau217 is higher in newborns than in AD BRAIN COMMUNICATIONS 2025, fcaf221 | 9

increased expression of adult over foetal isoforms.2 The vari­

ability in the decline rates and initial levels may be influenced
Funding
by differences in perinatal factors, as observed in cohort-2. K.B. is supported by the Swedish Research Council
The markedly higher p-tau217 levels with no aggregation or (#2017-00915 and #2022-00732), the Swedish Alzheimer
tangle formation in newborns compared to Alzheimer’s disease Foundation (#AF-930351, #AF-939721, #AF-968270 and
suggest distinct physiological processes governing tau phos­ #AF-994551), Hjärnfonden, Sweden (#ALZ2022-0006
phorylation during development versus neurodegenerative dis­ and #FO2024-0048-TK-130), the Swedish state under the
ease.2 It is still unknown if different p-tau markers show agreement between the Swedish government and the
different increases and trajectories in newborns like they do in County Councils, the ALF-agreement (#ALFGBG-965240
Alzheimer’s disease. Our findings suggest a unique regulatory and #ALFGBG-1006418), the European Union Joint
role for p-tau in neurodevelopment that warrants further inves­ Program for Neurodegenerative Disorders (JPND2019-
tigation.2,34 While our results show clear increases in blood-

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

466-236), the Alzheimer’s Association 2021 Zenith Award
based p-tau217 levels in both healthy and premature newborns, (ZEN-21-848495), the Alzheimer’s Association 2022-2025
the specific contributions of foetal tau versus adult tau isoforms Grant (SG-23-1038904 QC), La Fondation Recherche
to the detected blood levels remain unclear, as tau immunoas­ Alzheimer (FRA), Paris, France, the Kirsten and
says target regions shared by both foetal and adult isoforms. Freddy Johansen Foundation, Copenhagen, Denmark and
Understanding the mechanisms of maintaining high p-tau levels Familjen Rönströms Stiftelse, Stockholm, Sweden. The pro­
without triggering aggregation or pathology during foetal devel­ ject was funded by the Norwegian Research Council, JPND/
opment could offer valuable insights for Alzheimer’s disease and PMI-AD (NRC 311993), the National Institute of Health
other tauopathies.35 The natural protective mechanisms in new­ (R01 AG083874-01). F.G.-O. was funded by the Anna
borns may be mimicked as therapeutic approaches aiming at Lisa and Brother Björnsson’s Foundation and Emil och
modulating tau phosphorylation, improving tau clearance and Maria Palms Foundation. M.S.-C. receives funding from
preventing pathological tau aggregation in neurodegenerative the European Research Council (ERC) under the European
diseases. Future studies exploring the individual contributions Union’s Horizon 2020 research and innovation programme
of the different tau isoforms to the elevated phosphorylated (grant agreement no. 948677), Project ‘PI19/00155’, funded
tau levels in blood and the protective mechanisms that prevent by Instituto de Salud Carlos III (ISCIII) and co-funded by the
aggregation could lead to novel interventions that replicate these European Union, and from a fellowship from ‘la Caixa’
developmental conditions in Alzheimer’s disease. Foundation (ID 100010434) and from the European
Union’s Horizon 2020 research and innovation programme
under the Marie Skłodowska-Curie grant agreement no.
Strengths and limitations 847648 (LCF/BQ/PR21/11840004). B.-E.K. was supported
by a grant from Helse-Nord (HNF1540-20). H.Z. is a
This is the first report comparing levels of plasma p-tau217 in
Wallenberg scholar and a distinguished professor at the
newborns, controls and Alzheimer’s disease using currently
Swedish Research Council and receives funding from the
available blood-based immunoassays. Strengths of this study
Swedish Research Council (#2023-00356, #2022-01018
include the multicentric evaluation of blood-based p-tau217
and #2019-02397), the European Union’s Horizon Europe
and the inclusion of different age groups and two large cohorts
research and innovation programme under grant agreement
with newborn samples. However, there are important limita­
no. 101053962 and Swedish State Support for Clinical
tions that should be acknowledged, including the lack of lon­
Research (#ALFGBG-71320).
gitudinal plasma p-tau217 data and the lack of comparisons
with other biomarkers across all the cohorts included in this
study. Additionally, future studies should corroborate our re­
sults by using different immunoassays and platforms apart Competing interests
from Simoa. Our study provides an impetus to explore phos­ M.T. and P.H. are employees of Bioventix. HZ has served at sci­
phorylated tau biomarkers in cord blood. entific advisory boards and/or as a consultant for AbbVie Inc,
Acumen, Alector, Alzinova, ALZpath, Amylyx, Annexon,
Apellis, Artery Therapeutics, AZTherapies, Cognito
Supplementary material Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp,
Supplementary material is available at Brain Communications Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage
online. Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey
Labs, reMYND, Roche, Samumed, Siemens Healthineers,
Triplet Therapeutics and Wave, has given lectures sponsored
Acknowledgements by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, Lilly,
Novo Nordisk, Roche and WebMD and is a co-founder of
The authors would like to express their most sincere grati­ Brain Biomarker Solutions in Gothenburg AB (BBS), which is
tude to the cohort participants and relatives without whom a part of the GU Ventures Incubator Program (outside submit­
this research would not have been possible. ted work). K.B. has served as a consultant and at advisory
10 | BRAIN COMMUNICATIONS 2025, fcaf221 F. Gonzalez-Ortiz et al.

boards for AbbVie, AC Immune, ALZpath, AriBio, cytopathology in Alzheimer’s disease. Acta Neuropathol. 2002;
Beckman-Coulter, BioArctic, Biogen, Eisai, Lilly, Moleac Pte. 103(1):26-35.
6. Gonzalez-Ortiz F, Kac PR, Brum WS, Zetterberg H, Blennow K,
Ltd, Neurimmune, Novartis, Ono Pharma, Prothena,
Karikari TK. Plasma phospho-tau in Alzheimer’s disease:
Quanterix, Roche Diagnostics, Sanofi and Siemens Towards diagnostic and therapeutic trial applications. Mol
Healthineers; has served at data monitoring committees for Neurodegener. 2023;18(1):18.
Julius Clinical and Novartis; has given lectures, produced edu­ 7. Rajbanshi B, Guruacharya A, Mandell JW, Bloom GS.
cational materials and participated in educational programmes Localization, induction, and cellular effects of tau phosphorylated
at threonine 2171. Alzheimers Dement. 2023;19(7):
for AC Immune, Biogen, Celdara Medical, Eisai and Roche
2874-2887.
Diagnostics; and is a co-founder of Brain Biomarker 8. Nanoscale imaging of pT217-tau in aged rhesus macaque entorh­
Solutions in Gothenburg AB, which is a part of the GU inal and dorsolateral prefrontal cortex: Evidence of interneuronal
Ventures Incubator Program, outside the work presented in trafficking and early-stage neurodegeneration - Datta - 2024 -

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

this paper. M.S.-C. has served as a consultant and at advisory Alzheimer’s & Dementia - Wiley Online Library. Accessed
December 2, 2024. https://alz-journals.onlinelibrary.wiley.com/
boards for Roche Diagnostics International Ltd and Grifols
doi/10.1002/alz.13737
S.L., has given lectures in symposia sponsored by Roche 9. Islam T, Hill E, Abrahamson EE, et al. Phospho-tau serine-262 and
Diagnostics and Roche Farma, S.A. and was granted with a pro­ serine-356 as biomarkers of pre-tangle soluble tau assemblies in
ject funded by Roche Diagnostics International Ltd; payments Alzheimer’s disease. Nat Med. 2025;31(2):1-15.
were made to the institution (Barcelona Beta Research Center). 10. Brion JP, Octave JN, Couck AM. Distribution of the phosphory­
lated microtubule-associated protein tau in developing cortical neu­
B.-E.K. has served as a consultant for Biogen and advisory
rons. Neuroscience. 1994;63(3):895-909.
board for Eisai. T.F. has served as a consultant and at the advis­ 11. Schneider A, Biernat J, von Bergen M, Mandelkow E, Mandelkow
ory boards for Biogen, Novo Nordisk, Eli Lilly and Roche. The EM. Phosphorylation that detaches tau protein from microtubules
other authors declare no competing interest. (Ser262, Ser214) also protects it against aggregation into
Alzheimer paired helical filaments. Biochemistry. 1999;38(12):
3549-3558.
12. Mattsson N, Sävman K, Osterlundh G, Blennow K, Zetterberg H.
Converging molecular pathways in human neural development
Data availability and degeneration. Neurosci Res. 2010;66(3):330-332.
13. Gonzalez-Ortiz F, Ferreira PCL, González-Escalante A, et al.
No new software and/or algorithms, in-house scripts or pro­
A novel ultrasensitive assay for plasma p-tau217: Performance
gramme were generated to support this study. Requests for in individuals with subjective cognitive decline and early Alzheimer’s
the data sets used in the present study will be promptly disease. Alzheimers Dement. 2024;20(2):1239-1249.
reviewed by the corresponding authors and the University 14. Jonaitis EM, Janelidze S, Cody KA, et al. Plasma phosphorylated tau
of Gothenburg to verify whether the request is subject to 217 in preclinical Alzheimer’s disease. Brain Commun. 2023;5(2):
fcad057.
any intellectual property or confidentiality obligations.
15. Therriault J, Vermeiren M, Servaes S, et al. Association of phos­
Anonymized data can be shared by request from any quali­ phorylated tau biomarkers with amyloid positron emission tomog­
fied investigator for the sole purpose of replicating proce­ raphy vs tau positron emission tomography. JAMA Neurol. 2023;
dures and results presented in the article, if data transfer is 80(2):188-199.
in agreement with EU legislation. All requests for code 16. Jack CR, Andrews JS, Beach TG, et al. Revised criteria for diagnosis
and staging of Alzheimer’s disease: Alzheimer’s Association
used for data analyses and data visualization will be prompt­
Workgroup. Alzheimers Dement. 2024;20(8):5143-5169.
ly reviewed by the corresponding author and the University 17. Therriault J, Ashton NJ, Pola I, et al. Comparison of two plasma
of Gothenburg. p-tau217 assays to detect and monitor Alzheimer’s pathology.
EBioMedicine. 2024;102:105046.
18. Palmqvist S. Discriminative accuracy of plasma phospho-tau217 for
Alzheimer disease vs other neurodegenerative disorders. JAMA.
References 19.
2020;324(8):772-781.
Gonzalez-Ortiz F, Kirsebom BE, Yakoub Y, et al. Associations be­
1. Abasi LS, Elathram N, Movva M, Deep A, Corbett KD, tween changes in levels of phosphorylated tau and severity of cogni­
Debelouchina GT. Phosphorylation regulates tau’s phase separation tive impairment in early Alzheimer disease. Neurology. 2025;
behavior and interactions with chromatin. Commun Biol. 2024; 104(11):e213676.
7(1):1-17. 20. Kumar A, Janelidze S, Stomrud E, Palmqvist S, Hansson O,
2. Parra Bravo C, Naguib SA, Gan L. Cellular and pathological func­ Mattsson-Carlgren N. β-Amyloid-dependent and -independent gen­
tions of tau. Nat Rev Mol Cell Biol. 2024;25(11):845-864. etic pathways regulating CSF tau biomarkers in Alzheimer disease.
3. Buchholz S, Zempel H. The six brain-specific TAU isoforms Neurology. 2022;99(5):e476-e487.
and their role in Alzheimer’s disease and related neurodegenera­ 21. Moscoso A, Villain N. 2024 AA criteria for Alzheimer’s disease
tive dementia syndromes. Alzheimers Dementia. 2024;20(5): diagnosis: Mainly anchored at Aβ not tau. Alzheimers Dement.
3606-3628. 2024;20(12):9079-9081.
4. Brion JP, Smith C, Couck AM, Gallo JM, Anderton BH. 22. Cousins KAQ, Shaw LM, Shellikeri S, et al. Elevated plasma phos­
Developmental changes in τ phosphorylation: Fetal τ is transiently phorylated tau 181 in amyotrophic lateral sclerosis. Ann Neurol.
phosphorylated in a manner similar to paired helical filament-τ 2022;92(5):807-818.
characteristic of Alzheimer’s disease. J Neurochem. 1993;61(6): 23. Gonzalez-Ortiz F, Dulewicz M, Ashton NJ, et al. Association of ser­
2071-2080. um brain-derived tau with clinical outcome and longitudinal change
5. Augustinack JC, Schneider A, Mandelkow EM, Hyman BT. Specific in patients with severe traumatic brain injury. JAMA Netw Open.
tau phosphorylation sites correlate with severity of neuronal 2023;6(7):e2321554.
Plasma p-tau217 is higher in newborns than in AD BRAIN COMMUNICATIONS 2025, fcaf221 | 11

24. Gonzalez-Ortiz F, Karikari TK, Taylor-Te Vruchte D, et al. Plasma 30. Fladby T, Pålhaugen L, Selnes P, et al. Detecting at-risk
phosphorylated-tau217 is increased in Niemann–Pick disease type Alzheimer’s disease cases. J Alzheimers Dis. 2017;60(1):
C. Brain Commun. 2024;6(6):fcae375. 97-105.
25. Ashton NJ, Moseby-Knappe M, Benedet AL, et al. Alzheimer dis­ 31. Najm S, Löfqvist C, Hellgren G, et al. Effects of a lipid
ease blood biomarkers in patients with out-of-hospital cardiac ar­ emulsion containing fish oil on polyunsaturated fatty acid profiles,
rest. JAMA Neurol. 2023;80(4):388-396. growth and morbidities in extremely premature infants:
26. Bentivenga GM, Gonzalez-Ortiz F, Baiardi S, et al. Clinical value of A randomized controlled trial. Clin Nutr ESPEN. 2017;20:
novel blood-based tau biomarkers in Creutzfeldt–Jakob disease. 17-23.
Alzheimers Dement. 2025;21(2):e14422. 32. Gonzalez-Ortiz F, Turton M, Kac PR, et al. Brain-derived tau: A no­
27. Abu-Rumeileh S, Scholle L, Mensch A, et al. Phosphorylated tau vel blood-based biomarker for Alzheimer’s disease-type neurode­
181 and 217 are elevated in serum and muscle of patients with generation. Brain. 2022;146(3):1152-1165.
amyotrophic lateral sclerosis. Nat Commun. 2025;16(1):2019. 33. Wegmann S, Biernat J, Mandelkow E. A current view on tau protein
28. Ashton NJ, Suárez-Calvet M, Karikari TK, et al. Effects of pre- phosphorylation in Alzheimer’s disease. Curr Opin Neurobiol.

Downloaded from https://academic.oup.com/braincomms/article/7/3/fcaf221/8158110 by guest on 16 September 2025

analytical procedures on blood biomarkers for Alzheimer’s patho­ 2021;69:131-138.
physiology, glial activation, and neurodegeneration. Alzheimers 34. Seubert P, Mawal-Dewan M, Barbour R, et al. Detection of phos­
Dement (Amst). 2021;13(1):e12168. phorylated Ser262 in fetal tau, adult tau, and paired helical filament
29. Payne E. Umbilical cord blood pTau217 and BD-tau are associated with tau. J Biol Chem. 1995;270(32):18917-18922.
markers of neonatal hypoxia: A prospective cohort study. medRxiv. 35. Congdon EE, Ji C, Tetlow AM, Jiang Y, Sigurdsson EM.
https://doi.org/10.1101/2024.12.20.24319360, 24 December 2024, Tau-targeting therapies for Alzheimer disease: Current status and
preprint: not peer reviewed. future directions. Nat Rev Neurol. 2023;19(12):715-736.