Tracking neurodegenerative changes in humans is key to learning more about Alzheimer’s disease progression. Recent studies have suggested that blood-based biomarkers may be useful in detecting AD pathology, and plasma levels of tau phosphorylated at threonine 181 have shown promise. But exactly what this biomarker predicts, in whom, and how early in disease progression, remain unclear.

  • Baseline levels and longitudinal changes in plasma p-tau181 associate with neurodegeneration and cognitive decline.
  • Phospho-tau181 robustly predicts AD progression.
  • It could be used to monitor clinical trials and response to AD drugs.

Researchers led by Michael Schöll from the University of Gothenburg, Sweden, wanted to examine this marker in more detail. They had three questions: How do baseline and longitudinal p-tau181 associate with future neurodegeneration; does p-tau181 provide the same information as the neurofilament light chain biomarker; and how do imaging markers of neurodegeneration explain the link between p-tau181 and cognitive decline? Their findings, published in the January 11 JAMA Neurology, confirmed that, in a large cohort, plasma p-tau181 starting levels and increases over time robustly predict neurodegeneration in brain regions that typically degenerate in AD, but only in those patients who tested positive for Aβ plaques. They also found p-tau181 to be more AD-specific than NfL, which tracked with neurodegeneration regardless of the presence of amyloid plaques. Interestingly, neurodegeneration as measured by brain imaging could not fully explain the correlation between plasma p-tau181 and cognitive decline, suggesting some other component is important in this regard. The authors did not measure neurofibrillary tangles by PET.

“This paper extends on our past findings,” said Niklas Mattsson-Carlgren from Lund University, Sweden. Last November, the scientists had reported that high p-tau181 levels at baseline predict cortical atrophy in AD patients, and longitudinally increased p-tau181 forecasted accelerated atrophy (Hansson et al., 2020). “The new work compares p-tau181 to NfL. This contributes new data to blood biomarker research,” Mattsson-Carlgren said.

Last year saw an explosion of research on p-tau181, including a paper supporting the idea that plasma p-tau181 levels correlate with both plaques and tangles, and can discriminate between people with AD, tauopathies, and controls (Mar 2020 news). In cerebral spinal fluid, this biomarker was also linked to early amyloidosis (Mar 2020 news). 

For the current work, first author Alexis Moscoso and colleagues studied 1,113 participants in ADNI. Of these, 378 were cognitively normal, 537 had mild cognitive impairment (MCI), and 198 AD dementia. Participants had at least one FDG PET or structural MRI scan collected each time blood was taken. Blood samples were taken for up to eight years and stored; blood p-tau181 was measured in 2020.

First, Moscoso and colleagues determined how baseline plasma p-tau181 predicted progression. Using the FDG PET and MRI images, they found that, in symptomatic participants, higher p-tau181 at baseline came with faster decline in neuronal metabolism and atrophy in the temporoparietal regions that are characteristically involved in AD-related neurodegeneration, including the hippocampus. In cognitively normal participants, plasma p-tau181 also associated with future atrophy, but not hypometabolism, in AD-vulnerable regions. The plasma marker also predicted cognitive decline in all three groups, as measured by the Preclinical Alzheimer’s Cognitive Composite (PACC) or the Alzheimer’s Disease Assessment Scale-Cognitive Subscale (ADAScog). However, a mediation analysis indicated that neurodegeneration, as measured by hypometabolism or atrophy, explained less than half of the correlation between p-tau181 and cognition decline, implicating an as-yet-undefined factor, perhaps toxic forms of tau.

P-tau181 Predictions. Weakening brain metabolism (top), as seen by FDG PET scans, correlated with a year-on-year rise in plasma p-tau181 (lower graphs) in cognitively impaired people (right). The correlation was weaker among cognitively unimpaired participants (left). Yearly change in NfL levels also associated with hypometabolism, including in frontoparietal regions less affected by AD. [Courtesy of Moscoso et al., JAMA Neurology, 2021.]

The authors then investigated change in plasma p-tau181 over time. They found that increasing p-tau181 levels were linked to decreasing glucose metabolism and gray-matter volume among cognitively impaired participants.

Changes in NfL levels also correlated with progressive neurodegeneration in both unimpaired and impaired participants. Curiously, the shrinkage was predominantly in other frontoparietal regions less affected in AD (see image below). Deterioration in these frontoparietal regions is linked to age-related lesions in the brain caused by microvascular damage, suggesting that NfL is a general marker of neuronal degeneration (Habes et al., 2016). 

The authors also found that p-tau181’s association with hypometabolism, atrophy, and cognitive decline only held true in people who had a positive amyloid PET scan (see image below). According to the authors, this suggests that p-tau181, if indirectly, detects the accumulation of Aβ plaques over time.

Only if Aβ-Positive. Brain atrophy measured by MRI associated with plasma p-tau181 only in those cognitively unimpaired (left) and impaired (right) participants who were amyloid-positive, suggesting that p-tau181 reflects Aβ plaque accumulation. [Courtesy of Moscoso et al., JAMA Neurology, 2021.]

Taken together, these results add to a growing literature indicating plasma p-tau181 will be a valuable tool for estimating and monitoring AD progression. The authors believe this marker will be far easier to use on a wide array of participants, and cheaper, than CSF biomarkers and PET imaging. The builds on another longitudinal ADNI study the authors published last November, which linked plasma p-tau181 to Aβ pathology, tau PET, and p-tau181 in CSF (Moscoso et al., 2020). “With ADNI, we have this harmonized, standardized dataset and this abundance of biomarker data, so we can really shed light on a number of very important aspects of this novel blood biomarker,” said Schöll.

That said, correlations between p-tau181 and neurodegeneration were weaker in cognitively unimpaired than impaired participants. This begs the question: How good might plasma p-tau181 be as a prognostic tool for AD?

“Phospho-tau181 doesn’t look like a dream scenario when it comes to very early sensing of disease. We might need another marker, such as p-tau231, which from preliminary data may be more sensitive at early stages,” said Schöll. “Then for monitoring Alzheimer’s, you use 181, or maybe a combination of these. We don’t know yet.”

Plasma p-tau217 looks to be even more promising (Jul 2020 news). Curiously, a study published last month reported that this marker did not associate with atrophy in the hippocampus in amyloid-positive, symptomatic people (Mattsson-Carlgren et al, 2020). However, the lead author, Oskar Hansson of Lund University, cautioned that subgroup analyses were a very small part of the overall study and had low statically power. “We also need to remember that there were different assays used in the JAMA Neurology and Brain studies,” he told Alzforum. “To truly know that p-tau217 is a better biomarker than p-tau181, one need to compare identical assays, preferably mass-spectrometry-based (without use of different antibodies) to know if differences only have to do with the phospho-epitope or other assay-related factors, he said.

“This shouldn’t be a race to find the better marker,” said Schöll. The scientists agree that discrepancies like these underscore the need to directly compare different markers. “We need a head-to-head comparison to see where the complimentary information lies,” said Schöll. Henrik Zetterberg, University of Gothenburg, and colleagues are starting this work through the Global Biomarker Standardization Consortium, a group of scientists, clinicians, and industry leaders convened by the Alzheimer’s Association who are working to ready investigational biomarkers for widespread clinical use.—Helen Santoro


  1. The study by Moscoso and colleagues adds to the ongoing literature demonstrating that plasma p-tau181 is a sensitive and specific marker for Alzheimer’s disease (AD) pathology and clinical diagnosis.

    This study, using ADNI data, nicely highlights the prognostic utility of this plasma marker. Higher plasma P-tau181 levels were not only associated with greater declines in cognition, brain hypometabolism, and atrophy among MCI and AD participants, but also among those who were cognitively unimpaired.   Thus, this work further demonstrates that plasma P-tau181 changes very early in the AD pathological process.

    In addition, longitudinal change in plasma P-tau181 corresponded with change in cognition, suggesting that it may be feasible to use this marker as a surrogate endpoint in the future.

    As a whole, plasma p-tau appears to be a very promising AD marker.  A future research step will be to compare the different p-tau assays (e.g., SiMoA and MSD) and isoforms (e.g., p-tau181 and p-tau217) to determine which is best for a specific context of use (diagnostic vs. prognostic) or stage of the disease.

  2. Using longitudinally collected blood samples and data from the ADNI study from over 1,000 participants, Moscoso et al. evaluate the performance of p-tau181 as a biomarker of neurodegeneration and provide NfL for comparison.

    In the cognitively impaired group, the authors found p-tau181 showed associations with both longitudinal hypometabolism (FDG-PET, AD-typical Meta-ROI) and atrophy (vMRI, AD-signature ROI). The associations were numerically higher when compared with those observed for NfL. In a follow-on analysis with the groups further stratified by the presence of amyloid pathology, the authors find the association with hypometabolism and atrophy are dependent on the presence of amyloid pathology.

    Voxel wise associations performed by the authors illustrate subtle differences in the relationship between the blood biomarkers and imaging markers of neurodegeneration providing further support for the AD specificity of p-tau181 and the nonspecific nature of NfL. The mediation analysis conducted indicate the association of p-tau181 with cognitive decline is not solely due to neurodegeneration, as only  25-45 percent of cognitive decline is attributed to imaging derived neurodegeneration markers.

    Using this single large study, the authors provide further confirmation that p-tau181 is an Alzheimer’s disease specific biomarker that is associated with all key elements of AD (Amyloid, tau, neurodegeneration, and cognitive decline).  By incorporating NfL along with the neurodegeneration imaging markers as comparators, the authors are able to demonstrate subtle differences between the markers that support the specificity or lack of specificity for AD.

    Future studies exploring the influence of co-morbidities on both these plasma markers in large population-based studies may provide additional insight into factors that may lead to divergent biomarker results or aid in interpreting therapeutic response on biomarker outcomes in future studies.

Make a Comment

To make a comment you must login or register.


News Citations

  1. A Phospho-Tau Plasma Assay for Alzheimer’s?
  2. Different CSF Phospho-Taus Match Distinct Changes in Brain Pathology
  3. Plasma p-Tau217 Set to Transform Alzheimer’s Diagnostics

Series Citations

  1. The Alzheimer's Disease Neuroimaging Initiative

Paper Citations

  1. . Plasma phosphorylated tau181 and neurodegeneration in Alzheimer's disease. Ann Clin Transl Neurol. 2021 Jan;8(1):259-265. Epub 2020 Nov 29 PubMed.
  2. . White matter hyperintensities and imaging patterns of brain ageing in the general population. Brain. 2016 Apr;139(Pt 4):1164-79. Epub 2016 Feb 24 PubMed.
  3. . Time course of phosphorylated-tau181 in blood across the Alzheimer's disease spectrum. Brain. 2021 Feb 12;144(1):325-339. PubMed. Correction.
  4. . Longitudinal plasma p-tau217 is increased in early stages of Alzheimer's disease. Brain. 2020 Dec 5;143(11):3234-3241. PubMed.

Further Reading

No Available Further Reading

Primary Papers

  1. . Longitudinal Associations of Blood Phosphorylated Tau181 and Neurofilament Light Chain With Neurodegeneration in Alzheimer Disease. JAMA Neurol. 2021 Apr 1;78(4):396-406. PubMed.