While numerous fluid markers flag brain Aβ pathology in Alzheimer’s disease, markers of neurofibrillary tangles are few and far between. Now, scientists led by Oskar Hansson at Lund University in Sweden and Juan Lantero-Rodriguez of the University of Gothenburg in Sweden describe a new marker for tangles: N-terminal tau fragments, dubbed NTA-tau. In the February 17 Molecular Neurodegeneration, they report that people with plaques and tangles had twice as much plasma NTA-tau as did those with neither. It was the tangles that almost exclusively drove the plasma tau signal, the authors found, and only in AD. People with other tauopathies had normal plasma NTA-tau. Furthermore, high baseline levels of this marker predicted worse tangle accumulation, cortical atrophy, and cognitive decline over six years.

  • A new immunoassay detects N-terminal tau fragments.
  • Dubbed NTA-tau, these correlate with tangles in AD.
  • The assay identified tau PET-positive people with 80 percent accuracy.
  • High baseline NTA-tau predicted worse tangles, atrophy, and cognitive decline over six years.

“This work once again underscores the complexity of soluble tau forms in disease,” wrote Jeff Dage, Indiana University, Indianapolis (comment below). “Moreover, it adds evidence that exploring post-translational modifications, such as differential proteolytic cleavage, offers additional opportunities for novel biomarker discovery for Alzheimer’s disease and dementias.”

AD pathology is defined by fluid or imaging markers under the amyloid-tau-neurodegeneration (ATN) framework, whereby cerebrospinal fluid phosphotau-181 or tau PET reflect tau proteinopathy. PET scans are more onerous and expensive than plasma analysis, while p-tau181 tracks amyloid pathology as well, if not better, than tangles (Apr 2020 conference news; Sep 2022 conference news; Dec 2022 conference news).

In search of a good tangle marker, co-first authors Lantero-Rodriguez and Gemma Salvadó at Lund and colleagues developed two single-molecular array (Simoa) assays, A and B. They use two antibodies: one against amino acids 159 to 163 to capture tau, the other to N-terminal residues 6 to 18 for detection. The assays capture any tau containing residues 6 to 163, including full-length protein and N-terminal fragments. Lantero-Rodriguez told Alzforum that assay B worked poorly in blood. The A assay picked up N-terminal fragments in the CSF and plasma that correlated with tau PET in 130 amyloid-positive people across the AD spectrum (Snellman et al., 2022; Lantero-Rodriguez et al., 2023).

To test NTA-tau in a larger population, the scientists analyzed CSF and blood samples, PET and MRI scans, and cognitive test scores from 2,226 older adults in the Swedish BioFINDER-1 and BioFINDER-2 cohorts. Of these, 961 were cognitively normal, 354 were amyloid-positive based on CSF, 303 had mild cognitive impairment, 189 had AD dementia, and 419 had been diagnosed with non-AD MCI or dementia. Participants were around age 70; half were women.

Plasma NTA-tau seems to specifically reflect tangles associated with AD. It only increased in people who tested positive for amyloid and for tangles, but not in people who had only tangles (image below). Plasma NTA-tau more accurately correlated with tau PET than with amyloid PET, with areas under the curve of 0.80 and 0.67, respectively. Statistically, tangles, not amyloid or atrophy, explained most of the NTA-tau signal.

At later stages of AD, plasma NTA-tau had doubled (image below). It strongly associated with tangles, and atrophy, in the temporoparietal lobe, an area that accumulates tau during later stages of AD. People with high plasma NTA-tau did poorly on the modified preclinical Alzheimer’s cognitive composite (mPACC) and the Mini-Mental State Exam (MMSE).

Tagging Tangles. While NTA-tau was found in the plasma of amyloid-negative (A-) and tangle-negative (T-) controls (gray, left), levels were slightly higher in A+T- cases (blue, left), and highest in the A+T+ group (green, left). The A-T+ group (yellow, left) had no more plasma NTA-tau than controls. The same trend was seen across the clinical spectrum (right). From A- cognitively unimpaired controls (CU-) through to AD, NTA-tau rose. A- people diagnosed with non-AD dementia had the lowest NTA-tau, but it was higher in the A+ non-AD group. [Courtesy of Lantero-Rodriguez et al., Molecular Neurodegeneration, 2024.]

NTA-tau predicted faster disease progression. Over six years, people with the most plasma NTA-tau accumulated the most tangles in their temporal lobes, had the most cortical atrophy, and slipped the fastest on the mPACC or MMSE (image below). A faster rise in plasma NTA-tau correlated with worse atrophy and cognitive decline. Lantero-Rodriguez and Salvadó think plasma NTA-tau could serve as a screening tool to flag people likely to progress quickly in clinical trials, or as a secondary outcome measure to see how anti-amyloid immunotherapies affect tangles.

NTA-tau Up, Cognition Down. People with the most baseline plasma NTA-tau (purple) accumulated more tangles (top left), had worse cortical atrophy (top right), and slipped faster on the MMSE (bottom left) or mPACC (bottom right) over four to six years than did people with average (green) or low (yellow) plasma NTA-tau. [Courtesy of Lantero-Rodriguez et al., Molecular Neurodegeneration, 2024.]

“This [NTA-tau] assay appears to perform similarly to other new fluid markers postulated to better represent tau tangles, such as MTBR-tau243,” noted Tharick Pascoal of the University of Pittsburgh. Salvadó and Kanta Horie of Washington University in St. Louis previously reported that this chunk of tau’s microtubule-binding region ticked up in the CSF when tangles formed, and it tracked with waning cognition almost as well as did tau PET (Dec 2022 conference news; Aug 2023 conference news). MTBR-tau243 has not yet been measured in plasma.

In blood, another N-terminal tau fragment called NT-1, which spans residues 6 to 198, also closely correlated with tangle accumulation, neurodegeneration, and cognitive decline in AD (Dec 2020 news). That study's first author, Jasmeer Chhatwal at Massachusetts General Hospital, Boston, was encouraged that Lantero-Rodriguez and Salvadó saw similar correlations with plasma NTA-tau. “It makes me more confident about the underlying biology,” he told Alzforum.

Still, neither Chhatwal nor Horie are convinced that N-terminal tau, whether NTA-tau or NT1, directly measures insoluble tau, because these fragments do not end up in neurofibrillary tangles. “Things that correlate in time are not always driven by the same biological processes,” Chhatwal said. Horie noted that MTBR-tau243 is enriched in tangles, hinting it might be a direct marker (comment below).—Chelsea Weidman Burke


  1. Lantero-Rodriguez and colleagues have conducted a comprehensive characterization of the NTA-tau assay, which employs anti-tau mAb (HT7) for capture and anti-tau (Tau12) for detection. This assay was developed and validated by the Clinical Neurochemistry Laboratory in Sweden. Through two studies, BioFINDER-1 and BioFINDER-2, they have demonstrated associations with Amyloid PET, Tau PET, and Neurodegeneration (cortical thickness) in various ways. They identify a stronger association with tau PET than with amyloid in amyloid-positive subjects.

    These findings are largely consistent with our 2018 findings (Mielke et al. 2018) evaluating plasma tTau and P-tau181 versus amyloid PET, tau PET, and cortical thickness.   

    The authors acknowledge that this is not the first NTA-tau assay. It would be beneficial to compare this assay with the one developed by my team at Lilly (Ashton et al. 2022), where the epitopes are even closer together. It would also be worthwhile to examine these assays in clinical trial samples from amyloid removal therapy trials to compare the different responses between tTau and P-tau, thereby further investigating the differential relationship with amyloid.

    Interestingly, the levels in BioFINDER-1 differ significantly from those in BioFINDER-2, as shown in Table 1. Given the similarities in results, this bias might be uniform and due to pre-analytical differences between the studies. However, further investigation into this observation is warranted.

    This work once again underscores the complexity of soluble tau forms in disease. Moreover, it provides additional evidence that exploring post-translational modifications, such as differential proteolytic cleavage, offers additional opportunities for novel biomarker discovery for Alzheimer’s disease and related dementias.


    . Plasma phospho-tau181 increases with Alzheimer's disease clinical severity and is associated with tau- and amyloid-positron emission tomography. Alzheimers Dement. 2018 Aug;14(8):989-997. Epub 2018 Apr 5 PubMed.

    . Plasma and CSF biomarkers in a memory clinic: Head-to-head comparison of phosphorylated tau immunoassays. Alzheimers Dement. 2022 Nov 12; PubMed.

  2. This study tests the performance of another blood test from the UGOT team in the well-characterized Biofinder cohort. The authors showed convincing results that support that the variance of their new NTA-tau assay is much better explained by tau PET than by amyloid PET levels.

    Interestingly, the assay characteristics, when correlated with other biomarkers, were similar to what has been demonstrated for tau PET in previous studies. For example, NTA-tau levels increased across the disease spectrum but were particularly elevated in later stages. Its longitudinal changes were associated with Aβ status in asymptomatic and symptomatic and parallel atrophy and cognitive dysfunction. This new assay appears to perform similarly to other new fluid markers postulated to better represent tau tangles, such as MTBR-tau243.

  3. As most phosphorylated-tau biomarkers available in blood or CSF reflect mainly Aβ-related pathology, there is an urgent need for the identification of cost-effective and scalable fluid biomarkers specific for tau aggregate pathology in Alzheimer’s disease. Plasma NTA-tau discussed in this paper increased across the AD continuum, especially during mid- to late-AD stages, and was associated with brain tau aggregates in terms of tau-PET. These findings will make waves in the AD biomarker research field.

    Here, it is important to understand the biology behind the characteristics of the biomarker, i.e., why plasma NTA-tau increases in the late stage of AD. The authors have discussed the possibility that plasma NTA-tau may serve as an indicator of AD tau pathology by showing a correlation between plasma NTA-tau and tau-PET, but whether this relationship is truly causal requires careful attention. Since the NTA-tau is unlikely enriched in neurofibrillary tangles (NFT), it is possible that NTA-tau is not released directly from NFT but is increased in extracellular spaces by passive secretion of soluble tau due to neurodegeneration induced by NFTs accumulation. The exact stage in which plasma NTA-tau levels are altered should be further investigated.

    We have recently reported CSF MTBR-tau243 as a specific biomarker for AD tau pathology through validation by large clinical cohorts analyses (Horie et al., 2023). The MTBR-tau243 fragment may have a direct causal relationship with AD tau pathology, since this tau species has been found to be significantly enriched in AD brain tau tangles (Horie et al., 2021; Wesseling et al., 2020). 

    Regarding plasma NTA-tau, it is not only a potential biomarker reflecting AD tau pathology, but also a promising biomarker for subsequent neurodegeneration. Recently, we proposed a strategy for staging AD by measuring a panel of fluid biomarkers from one aliquot of CSF (Salvadó et al., 2023). The biomarker panel includes Aβ42/40 and p-tau217 for early continuum prediction, p-tau205 and MTBR-tau243 for mid- to late-stage continuum, and total tau for the latest neurodegeneration stage. In the future, plasma NTA-tau may be promising as a marker of neurodegeneration stage to construct the plasma biomarker panel.


    . CSF MTBR-tau243 is a specific biomarker of tau tangle pathology in Alzheimer's disease. Nat Med. 2023 Aug;29(8):1954-1963. Epub 2023 Jul 13 PubMed.

    . Corrigendum to: CSF tau microtubule binding region identifies tau tangle and clinical stages of Alzheimer's disease. Brain. 2021 Oct 22;144(9):e82. PubMed.

    . Tau PTM Profiles Identify Patient Heterogeneity and Stages of Alzheimer's Disease. Cell. 2020 Dec 10;183(6):1699-1713.e13. Epub 2020 Nov 13 PubMed.

    . Novel CSF tau biomarkers can be used for disease staging of sporadic Alzheimer's disease. 2023 Jul 16 10.1101/2023.07.14.23292650 (version 1) medRxiv.

  4. The characterization of tau in CSF and blood has marked a milestone in the field of biomarkers for AD, and we now have a range of assays targeting different tau forms. However, most tau biomarkers are highly correlated among each other and this prompts the question: Do they provide unique insights into the disease?

    In my view, three critical aspects need to be considered. Firstly, Alzheimer's as a continuum involves sequential biomarker changes. We know that changes in p-tau231 and p-tau217 occur early, while p-tau235 changes in later stages of preclinical AD (Lantero‐Rodriguez et al., 2021; Milà-Alomà et al., 2022). Second, we realized that the "T" in the ATN classification not only reflects soluble tau changes (mainly driven by Aβ), but also insoluble tau deposition. The former may be more related to Aβ biomarkers, while the latter to Tau PET. Lastly, beyond early detection, there is a need for biomarkers that can effectively stage and monitor the disease's progression across all the stages of the disease.

    An innovative NTA-tau assay by Rodriguez-Lantero and Salvadó et al. offers additional information, particularly in intermediate and later disease stages. Plasma NTA-tau is characterized by its increase throughout the continuum of AD but, unlike other tau biomarkers, that increase mostly occurs in later stages of the disease (as shown by its higher levels in the A+T--, Braak V-VI, and A+MTL+N++ groups). Plasma NTA-tau variation is mainly explained by Tau PET, not amyloid pathology biomarkers, suggesting that it reflects tau insoluble aggregates. Moreover, plasma NTA-tau predicts tau accumulation, brain atrophy and cognitive decline, and longitudinal changes in plasma NTA-tau are correlated with disease progression.

    Plasma NTA-tau is not the optimal plasma biomarker for detecting AD, neither in early nor later stages, yet it offers valuable insights. Firstly, it aids in disease staging, particularly tracking insoluble tau deposition. This has therapeutic implications, as certain novel disease-modifying treatments may be indicated at specific stages. Additionally, in intermediate and late disease stages, plasma NTA-tau serves as a prognostic biomarker, supporting clinical trials, and may provide prognosis information to patients. Longitudinal changes in plasma NTA-tau also suggest its potential use in monitoring insoluble tau deposition.

    An essential consideration involves comparing plasma NTA-tau with other tau biomarkers related to tau pathology (such as MTBR-tau234), markers of neurodegeneration linked to AD pathology (like BD-tau or t-tau), or those independent of AD pathology (such as NfL).

    Another crucial aspect is the apparent specificity of plasma NTA-tau for AD. As observed in most tau biomarkers, increases in CSF NTA-tau have been reported during acute or rapidly progressive neurological diseases like stroke or CJD (Snellman et al., 2022). Investigating the capacity of plasma NTA-tau to indicate neurodegeneration beyond AD is important.

    In conclusion, plasma NTA-tau emerges as a promising fluid biomarker for in vivo tracking of tau aggregation, and could be used for disease staging and progression monitoring.

Make a Comment

To make a comment you must login or register.


News Citations

  1. 217—The Best Phospho-Tau Marker for Alzheimer’s?
  2. Head-to-Head Study Confirms Plasma p-Tau231 Rises First in Early AD
  3. Plasma P-tau217 Picks Up Plaques, Tangles, Future Decline
  4. New Data Bolsters MTBR-Tau243 as Fluid Marker for Tangles
  5. CSF MTBR-tau-243 Tracks Tangles, Plummets in Response to Antibody
  6. Plasma NT1: This Tau Snippet Predicts Cognitive Decline in Alzheimer’s

Paper Citations

  1. . N-terminal and mid-region tau fragments as fluid biomarkers in neurological diseases. Brain. 2022 Aug 27;145(8):2834-2848. PubMed.
  2. . Plasma and CSF concentrations of N-terminal tau fragments associate with in vivo neurofibrillary tangle burden. Alzheimers Dement. 2023 May 15; PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Plasma N-terminal containing tau fragments (NTA-tau): a biomarker of tau deposition in Alzheimer's Disease. Mol Neurodegener. 2024 Feb 17;19(1):19. PubMed.