Elusive as it has been for years, the goal of a blood test to detect Alzheimer’s disease appears suddenly achievable. First came the debut of two mass spectrometry methods and an ELISA that quantified Aβ in the blood reliably enough to pick out people with brain amyloidosis (Feb 2018 news; Dec 2017 conference newsJul 2017 conference news). And now, researchers at the first Advances in Alzheimer’s and Parkinson’s Therapies Focus Meeting, held March 15–18 in Turin, Italy, have brought plasma p-tau181 into play, as well.

  • Two independent groups report that plasma p-tau181 tells people with prodromal and symptomatic AD apart from healthy controls.
  • The biomarker accurately predicts brain amyloidosis.
  • The findings bring a blood test for AD closer to reality.

Dubbed AAT-AD/PD, the new conference merges the International Geneva/Springfield Symposium on Advances in Alzheimer Therapy with the International Conference on Alzheimer’s and Parkinson’s Diseases. It will be held every two years, alternating with AD/PD. The inaugural meeting convened 1,336 attendees from 64 countries to the former Fiat main factory building, which was reconfigured into a congress center in the early 1990s and still has a racetrack on its roof.

In Turin, researchers led by Charles S.Y. Yang at the biotech company MagQu Co., based in New Taipei City, Taiwan, presented a method that uses magnetic spin to detect minute quantities in blood of a given target protein, in this case tau phosphorylated at position 181. This enabled detection of consistent differences between healthy controls, people with prodromal AD, and those with dementia, Yang reported. Meanwhile, Michelle Mielke at the Mayo Clinic in Rochester, Minnesota, in collaboration with Jeffrey Dage and colleagues at Eli Lilly and Co., Indianapolis, described an ELISA-based method for quantifying plasma p-tau181 that also distinguished these diagnostic groups.

The rooftop racetrack on former FIAT manufacturing building, site of the AAT-AD/PD conference. [Photo: André P. Meyer-Vitali/Creative Commons.]

“Plasma p-tau181 appears promising, with an impressive area under the curve,” noted Kaj Blennow of the University of Gothenburg, Sweden, in an overview lecture on fluid biomarkers. Blennow is collaborating with Yang to validate his assay. Others at the meeting reacted to the data with enthusiasm. “This is what we’ve been waiting for,” said Hilkka Soininen of Kuopio University Hospital, Finland.

Few previous studies had examined plasma p-tau, although last year researchers reported that the marker distinguished people who had sustained head injuries from healthy controls better than did total tau in plasma. Likewise, Mielke and Dage found plasma total tau to be a dubious marker, reporting that its levels in healthy controls only weakly correlated with later cognitive decline and progression to AD (Sep 2017 news). Total tau is believed to reflect neuronal injury and neurodegeneration, whereas phosphorylated tau appears to be more specific for the pathology present in Alzheimer’s (Aug 2016 conference news). 

Because neurofibrillary tangles contain hyperphosphorylated tau, p-tau levels probably rise early in the pathogenesis of AD, before tangles form, Yang suggested at AAT-AD/PD. The question is, can it be detected specifically in the blood? This presents a challenge because, at a few pg/ml, its levels are extremely low, about 10 times lower than plasma t-tau. To overcome this, Yang used a technique developed by MagQu called immunomagnetic reduction (IMR). The researchers coat magnetic particles with antibodies against the protein of interest, then mix these particles with a blood sample. When p-tau181 binds to the particles, its mass slows down their rotation in a magnetic field (see image below and company webpage). This change in rotation can be detected with a powerful magnetic sensor, the superconducting quantum interference device (SQUID). Yang said that spinning the particles so fast that non-specific antibody binding is removed allows the researchers to quantify only the protein of interest.

Magnetic Detection.

Antibodies bound to magnetic particles capture p-tau181 in plasma. The bound tau slows the particles’ spin, allowing detection of tiny quantities. [Courtesy of Charles Yang.]

Yang and colleagues used IMR to compare p-tau181 levels in 23 healthy controls, 29 people with mild cognitive impairment due to AD, and 21 people with mild AD. The method measured rising levels, from an average of 2.46 pg/ml in controls, to 4.41 in MCI and 6.15 in AD. The difference was significant, with a p value of 0.001, Yang said.

The method has potential as a diagnostic test, Yang asserted. A cutoff of 3.1 pg/ml separated controls from MCI with an area under the curve of 86 percent, while a cutoff of 5.7 pg/ml separated MCI and AD with an AUC of 78 percent. Moreover, p-tau181 better detected AD progression than other plasma biomarkers did. Using the same method, t-tau and the Aβ42/Aβ40 ratio each distinguished healthy controls from MCI, but not MCI from AD, Yang reported. These data appeared January 23 in Alzheimer’s and Dementia (Yang et al., 2018). 

In Turin, the audience peppered Yang with questions. Some asked what forms of tau IMR detects; Yang replied that it cannot distinguish between tau monomers or oligomers. Others asked how plasma p-tau181 relates to other AD biomarkers, and Yang said his company is preparing studies in the U.S. and Japan to correlate p-tau181 with tau PET and serum neurofilament light (Nov 2017 news). He is currently repeating the findings in a Swedish cohort in collaboration with Blennow and colleagues.

For their part, Mielke and Dage took a different approach to measuring plasma p-tau. They used a sandwich immunoassay in which both antibodies against p-tau181 were monoclonals. This improved specificity over traditional ELISAs, Dage told Alzforum. He noted that the assay was able to detect p-tau181 in 97 percent of the tested plasma samples, with measurements varying about 8 percent between runs.

Following up on their earlier t-tau study, the researchers used this assay to compare plasma p-tau181 and t-tau levels in 172 healthy controls, 57 people with MCI, and 40 with AD participating in the Mayo Clinic Study on Aging. As did Yang and colleagues, they, too, found a statistically significant rise in plasma p-tau181 with advancing disease, from 5.1 pg/ml in controls to 5.7 in MCI and 11.8 in AD. Moreover, p-tau181 correlated with both amyloid and tau PET, Mielke reported in Turin. Plasma p-tau181 predicted the presence of brain amyloid with an accuracy of 75 percent. This is as good a predictor as using age and ApoE status, she noted. However, it is lower than current plasma Aβ tests, which detect amyloidosis with an accuracy of up to 90 percent. Mielke suggested p-tau181 could make a good screening measure for trials.

Plasma t-tau, on the other hand, correlated strongly with cortical atrophy but only weakly with tau PET. T-tau levels did not relate to a person’s diagnosis, and correlated with p-tau181 only among those who were amyloid-positive, Mielke said. Total tau in plasma predicted brain amyloid only 55 percent of the time, in line with previous findings that this marker flags degeneration and injury more generally, not AD pathology specifically.

Lilly is now replicating its p-tau and t-tau findings using samples from its clinical trials. In addition, the researchers plan to test the p-tau181 ELISA in non-AD tauopathies.—Madolyn Bowman Rogers


  1. We also reported the usefulness of plasma p-tau181 as a blood biomarker for Alzheimer’s disease and Down's syndrome (DS) (Tatebe et al., 2017).

    We have developed a novel ultrasensitive immunoassay to quantify plasma p-tau181 by using the Simoa system by Quanterix. With it, we measured the levels of plasma p-tau181 in three cohorts, including controls and patients with Alzheimer's disease and Down's syndrome. In the first cohort, composed of 20 AD patients and 15 age-matched controls, the plasma levels of p-tau181 were significantly higher in the AD patients than those in the controls (0.171 ± 0.166 pg/ml in AD versus 0.0405 ± 0.0756 pg/ml in controls, p = 0.0039).

    In the second cohort, composed of 20 patients with DS and 22 age-matched controls, the plasma concentrations of p-tau181 were significantly higher in the DS group (0.767 ± 1.26 pg/ml in DS versus 0.0415 ± 0.0710 pg/ml in controls, p = 0.0313). There was a significant correlation between the plasma levels of p-tau181 and age in the DS group (R2 = 0.4451, p = 0.0013). All of the DS individuals showing an extremely high concentration of plasma p-tau181 (> 1.0 pg/ml) were older than 40 years of age.

    In the third cohort, composed of eight AD patients and three patients with other neurological diseases, the levels of plasma p-tau181 significantly correlated with those of CSF p-tau181 (R2 = 0.4525, p = 0.023).

    Our exploratory pilot study warrants further large-scale and well-controlled studies to validate the usefulness of plasma p-tau181 as an urgently needed surrogate marker for the diagnosis and disease progression of AD. Now we are searching for research groups that can carry out collaborative work with us, especially work to determine whether or not there is a correlation between the levels of plasma p-tau and the extent of brain tau accumulation detected with tau-PET.


    . Quantification of plasma phosphorylated tau to use as a biomarker for brain Alzheimer pathology: pilot case-control studies including patients with Alzheimer's disease and down syndrome. Mol Neurodegener. 2017 Sep 4;12(1):63. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Closing in on a Blood Test for Alzheimer’s?
  2. Automated CSF Tests: Check. Blood Tests: In the Works
  3. Finally, a Blood Test for Alzheimer’s?
  4. Bloodborne Tau: Foggy Window into the Brain for TBI, Dementia
  5. Staging of Alzheimer’s, the Second: Neurodegeneration Does Not Equal Tauopathy
  6. Serum NfL Detects Preclinical AD, Reflects Clinical Benefit

Paper Citations

  1. . Assay of Plasma Phosphorylated Tau Protein (Threonine 181) and Total Tau Protein in Early-Stage Alzheimer's Disease. J Alzheimers Dis. 2018;61(4):1323-1332. PubMed.

External Citations

  1. company webpage

Further Reading