- Design
- Retrospective single-centre study with multivariable logistic regression
- Population
- 706 participants with clear plasma p-tau217 results and confirmatory amyloid PET or CSF biomarkers
- Primary outcome
- Discordance between plasma p-tau217 and reference biomarkers
- Effect
- 8.5% discordant; false negatives 5.6% vs false positives 2.8% (P=0.007); older age OR 1.61 per decade (1.23 to 2.14)
This retrospective Mayo Clinic study included 706 participants from memory and ageing studies with clearly positive or negative plasma p-tau217 and confirmatory amyloid PET or CSF p-tau181/amyloid-beta 42. Median age was 70.
Of 706, 60 (8.5%) were discordant. False negatives (5.6%) were more frequent than false positives (2.8%; P=0.007). Discordance was associated with older age (OR 1.61 per decade, 95% CI 1.23 to 2.14) and obstructive sleep apnoea (OR 1.94, 1.08 to 3.49).
False negatives, compared with true positives, were linked to milder clinical stage, higher MMSE scores, sleep apnoea, higher body mass index and higher eGFR. False positives, compared with true negatives, were associated with older age, thyroid disease, lower eGFR and advanced disease stage.
The study excluded intermediate plasma values and used research cohorts, so real-world discordance in an unselected clinic population may differ. The pattern fits known biology: renal clearance and volume of distribution can shift plasma concentrations.
- Interpret plasma p-tau217 alongside clinical stage, kidney function and body mass index.
- A negative result in a patient with mild symptoms and high BMI or eGFR can be a false negative; consider confirmatory testing.
- In older patients with reduced eGFR, a positive result may be a false positive.
- Confirm with amyloid PET or CSF before starting disease-modifying therapy where a plasma result is the only evidence.
- Record sleep apnoea and thyroid disease when interpreting results.
Why it matters
It identifies who is most at risk of a wrong blood-test answer when treatment decisions follow it.
Don't overread it
A single-centre retrospective study in research participants cannot define test accuracy in routine practice.
The statistics, in plain English
An odds ratio of 1.61 per decade means each extra ten years of age raised the odds of a discordant result by about 60%. The study excluded indeterminate results, so the 8.5% applies only to clearly positive or negative values.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for neurology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free