Blood biomarker correlates with imaging, CSF

A new assay for blood p-tau181 may be useful tool in Alzheimer’s disease diagnosis, management, and clinical trials, a mixed cohort study showed.

The assay, which had improved sensitivity for tau protein phosphorylated at threonine 181 (p-tau181), appeared to predict tau and amyloid-beta (Aβ) pathologies, differentiate Alzheimer’s disease from other neurodegenerative disorders, and identify Alzheimer’s disease across the clinical continuum, reported Kaj Blennow, MD, of the University of Gothenburg in Sweden and coauthors, in Lancet Neurology.

“The blood p-tau181 assay identified Alzheimer’s disease at the very early stages of disease and demonstrated high diagnostic accuracy, with stepwise increases across the Alzheimer’s disease continuum,” wrote Blennow and colleagues.

“Our blood p-tau181 assay appeared to be specific to Alzheimer’s disease, differentiating it from other neurodegenerative diseases with high accuracy,” they added. “Additionally, blood p-tau181 predicted cognitive decline and hippocampal atrophy over a period of 1 year, making it suitable as an Alzheimer’s disease progression marker.”

Blennow and coauthors evaluated their assay in four cohorts, some with imaging and cerebrospinal fluid (CSF) biomarker data. They found that measured p-tau181:

  • Increased across the continuum of cognitive impairment: lowest in Aβ-negative young adults and unimpaired older adults, higher in Aβ-positive unimpaired older adults and highest in Aβ-positive mild cognitive impairment (MCI) and Alzheimer’s disease groups.
  • Discriminated Alzheimer’s disease from cognitively normal adults and other neurodegenerative disorders, though in one cohort it did not differentiate Alzheimer’s disease from MCI.
  • Was associated with both PET-measured cerebral tau (area under curve [AUC] 83-93%) and PET-measured Aβ (AUC 76-88%).
  • Predicted 1-year cognitive decline (P=0.0015) and hippocampal atrophy (P=0.015) in a subset of one cohort with 1-year MRI and cognitive testing follow-up.

“The authors have shown the validity of their plasma p-tau181 assay in a rigorous and convincing manner,” noted Clifford Jack, Jr., MD, of the Mayo Clinic in Rochester, Minnesota, in an accompanying editorial. “This collective evidence suggests that plasma p-tau181 is an indicator of the pathophysiological state leading to both of the defining proteinopathies of Alzheimer’s disease, amyloid β deposition and deposits that contain a mixture of 3R and 4R tau isoforms,” Jack continued.

Alzheimer’s disease biomarkers include PET and CSF results, but have limited use outside research settings due to availability, cost, or invasiveness. Interest has grown in blood markers and p-tau species, including those phosphorylated at the 181 site and another candidate, p-tau217.

In their study, Blennow and colleagues evaluated their immunoassay in four cohorts:

  • A discovery cohort (n=37) that included Alzheimer’s patients with CSF biomarker data versus age-matched controls.
  • An observational research cohort TRIAD (n=226) that included younger patients and some with frontotemporal dementia in addition to a spectrum of cognitive impairment.
  • An observational research cohort BioFINDER-2 (n=763) that included participants with several other neurodegenerative disorders in addition to a spectrum of cognitive impairment.
  • A primary care cohort (n=105) that included community members without neurologic diagnosis and patients referred for specialist dementia evaluation prior to evaluation.

A subset of TRIAD participants that had 1-year MRI and cognitive testing (Mini-Mental State Examination; MMSE) follow-up.

Plasma and CSF p-tau181 levels in the discovery cohort showed correlation (r0.71, P<0.0001) and both plasma and serum p-tau181 accurately diagnosed Alzheimer’s disease (AUC 96% and 90%, respectively). In the TRIAD cohort, plasma p-tau181 was correlated with CSF levels of both p-tau181 and Aβ.

Evidence that p-tau181 may be informative across the Alzheimer’s spectrum was seen in the BIOFINDER-2 and primary care cohorts. In the latter, p-tau181 distinguished Alzheimer’s disease from young adults (AUC and accuracy=100%), cognitively unimpaired older adults (AUC 84.44%, accuracy >90%), but not from MCI (AUC 55.00%).

“The overlap between participants with MCI and Alzheimer’s disease in the primary care cohort is likely to be driven by patients with MCI already having Alzheimer’s disease dementia phenotypes, which cannot be excluded in this cohort without detailed PET or CSF biomarker data,” the researchers wrote.

Despite this, the fact that the assay distinguished clinically defined probable Alzheimer’s disease from young adults and cognitively unimpaired older adults “indicates applicability in a general medical setting,” the editorialist noted.

In the BIOFINDER-2 cohort, plasma p-tau181 distinguished Alzheimer’s disease from vascular dementia (AUC 92%), progressive supranuclear palsy or corticobasal syndrome (AUC 88%), behavioral variant frontotemporal dementia or primary progressive aphasia (AUC 83%), and Parkinson’s disease or multiple systems atrophy (AUC 82%).

Plasma p-tau181 also performed better than Alzheimer’s disease risk factors (age, APOE4 genotype, or both) and other plasma biomarkers in predicting Alzheimer’s disease diagnosis, increased tau PET, and increased Aβ PET. “The strong correlation between plasma p-tau181 and amyloid β PET, together with the increased plasma p-tau181 in amyloid β PET-positive and tau PET-negative (Braak 0) individuals suggests that this new test detects Alzheimer’s disease type pathology in the very early disease stages,” the authors noted.

TRIAD cohort participants with 1-year follow-up for MMSE (n=85) and MRI (n=88) showed a correlation between plasma p-tau181 and baseline (P<0.0001) and 1-year worsening (P=0.0015) on the MMSE, as well as baseline (P<0.0001) and 1-year change (P=0.015) in hippocampal atrophy.

Plasma p-tau181 may be useful in clinical trials and “could substantially increase efficacy and decrease cost of clinical trials by reducing or eliminating negative PET scans that are now commonly encountered in the screening process for disease modifying trials,” Jack observed. It could also enable research with a biological definition of Alzheimer’s disease where it has not been possible — in large epidemiological studies or in low income countries, he added.

Limitations of the study include heterogeneity in measurements and populations across cohorts. Studies comparing the diagnostic performance of plasma p-tau181 and p-tau217 will be important, Jack added.

  1. A new assay for plasma p-tau181 may be useful in Alzheimer’s disease diagnosis, management, and clinical trials, a mixed cohort study showed.

  2. The assay appeared to predict tau and amyloid-beta (Aβ) pathologies, differentiate Alzheimer’s disease from other neurodegenerative disorders, and identify Alzheimer’s disease across the clinical continuum.

Paul Smyth, MD, Contributing Writer, BreakingMED™

Funding was provided by the Alzheimer Drug Discovery Foundation, European Research Council, Swedish Research Council, Swedish Alzheimer Foundation, Swedish Dementia Foundation, and Alzheimer Society Research Program.

Blennow is a co-founder of Brain Biomarker Solutions in Gothenburg, a GU Ventures-based platform company at the University of Gothenburg.

Jack serves on an independent data monitoring board for Roche and has consulted for Eisai, but he receives no personal compensation from any commercial entity. He receives research support from the NIH and the Alexander Family Alzheimer’s Disease Research Professorship of the Mayo Clinic.

Cat ID: 33

Topic ID: 82,33,282,485,730,33,192