Which tau biomarkers are most informative in AD?

Aggregates of hyperphosphorylated and C-terminally truncated microtubule-associated tau protein are a feature in 80% of dementia disorders. An overview of novel and established tau biomarkers and their role in the detection, diagnosis, and prognosis of tau-associated neurodegenerative disorders in research and clinical trial settings was presented at AAIC 2019.

Increased entorhinal cortex tau is associated with worse memory

In cognitively normal and Alzheimer's disease (AD) individuals, the production rate of tau positively correlates with the amount of amyloid-beta (Aβ) plaques, indicating a biological link between Aβ plaques and tau physiology.1

The following Aβ and tau biomarkers are therefore used by the National Institute on Aging and Alzheimer’s Association research framework to define Alzheimer’s continuum:

  • elevated cerebrospinal fluid (CSF) phosphorylated tau (P-tau)
  • cortical tau positron emission tomography (PET) ligand binding
  • cortical Aβ PET ligand binding
  • low CSF Aβ42 or CSF Aβ42/Aβ42 ratio2


How informative is tau PET?

FTP ([18F]flortaucipir) PET tau correlates strongly with tau neuropathology3

Tau PET imaging in vivo reveals regional patterns and changes of tau pathology in the living human brain and how these are related to Aβ burden, cognitive decline, and neural activation.

Based on the results of published research, the presenter concluded that:

  • increased entorhinal cortex tau is associated with worse memory regardless of the presence of Aβ
  • very high entorhinal cortex tau levels are uncommon in Aβ-negative individuals
  • Aβ facilitates the increase of medial temporal lobe tau and tau spread and might potentiate tau-induced memory decline
  • rapid prospective memory decline is usually only seen in subjects with both Aβ and tau pathology

The audience was advised that although tau PET is a useful tool and enables the quantification of tau, off-target binding and other sources of noise can mean that quantification is challenging when the signal is low.4

In addition, although tau PET data suggest that tau pathology in aging, even without Aβ, relates to memory decline and neural alterations, it remains unclear whether this represents early Alzheimer's disease (AD) and which factors drive the spread of tau pathology in the progression from aging towards AD.5


How informative is CSF tau?

CSF tau biomarkers correlate weakly with tau neuropathology6

Total tau (T-tau) and phosphorylated tau (P-tau) are established CSF biomarkers for AD, the presenter explained:7

  • Increased CSF T-tau probably reflects neurodegeneration
  • Increased CSF P-tau probably reflects brain tau pathology and is only found in AD

The N-terminal and C-terminal of the tau protein provide different information, she added. CSF tau assays targeting specific fragments may therefore be useful for differential diagnosis and characterization of tauopathies.


Does Aβ induce changes in tau metabolism?

CSF tau measures are changed early in AD, possibly reflecting altered tau secretion and phosphorylation in response to Aβ pathology, the final presenter of the session explained. In contrast, 18F-FTP PET tau changes in parallel with cognitive decline and atrophy and is likely a more accurate measure of tau aggregation.

He suggested that Aβ pathology therefore induces changes in tau metabolism leading to the subsequent formation of AD-like tau aggregates.8

Our correspondent’s highlights from the symposium are meant as a fair representation of the scientific content presented. The views and opinions expressed on this page do not necessarily reflect those of Lundbeck.


  1. Sato C, et al. Neuron. 2018;97:1284–98.
  2. Jack C, et al. Alzheimers Dement 2018;14:535–62.
  3. Irwin D, et al. Ann Neurol 2017;82:247–58.
  4. Baker S. Abstract 31379. Presented at AAIC 2019 Los Angeles 17 July 2019.
  5. Maas A, et al. Abstract 31392. Presented at AAIC 2019 Los Angeles 17 July 2019.
  6. Smith R, et al. JAMA Neurol 2019;76:310–17.
  7. Blennow K, Zetterberg H. J Intern Med 2018;284:643–63.
  8. Mattsson N, et al. Abstract 31374. Presented at AAIC 2019 Los Angeles 17 July 2019.