Plasma p-tau217: one result in 12 disagreed with PET or CSF

Design

Retrospective diagnostic concordance study, single centre

Population

706 memory and ageing study participants with definite plasma p-tau217 results

Primary outcome

Discordance with amyloid PET or CSF p-tau181/Aβ42

Effect

8.5% discordant (5.6% false negative, 2.8% false positive); eGFR OR 1.63 per 10 mL/min for false negatives

Plasma p-tau217 is being adopted to diagnose Alzheimer pathology and to select patients for anti-amyloid therapy. This retrospective Mayo Clinic study took 706 research participants (median age 70) with clearly positive or clearly negative plasma results and compared them with amyloid PET or CSF.

8.5% were discordant. False negatives (5.6%) were twice as common as false positives (2.8%). After adjusting for age, false negatives were associated with earlier clinical stage, higher BMI and higher eGFR (OR 1.63 per 10 mL/min). False positives were associated with lower eGFR (OR 0.54 per 10 mL/min higher) and advanced dementia; thyroid disease and older age were associated in unadjusted analysis. Obstructive sleep apnoea was linked to discordance overall.

The pattern fits biology: reduced kidney clearance raises plasma p-tau, and larger body volume dilutes it. The practical change is that a plasma result should be read alongside kidney function, body size and clinical stage, and confirmed with PET or CSF before an irreversible or costly treatment decision. These were research participants with intermediate values excluded, so real-world discordance may be higher.

• Check eGFR before interpreting plasma p-tau217; reduced kidney function was linked to false positives.
• Interpret a negative result cautiously in early, mild symptoms or higher BMI.
• Confirm a plasma result with amyloid PET or CSF before starting anti-amyloid therapy.
• Consider thyroid disease and sleep apnoea when results do not fit the clinical picture.
• Remember this study excluded intermediate values, which are the hardest to interpret.

Why it matters

Blood tests are moving into memory clinics fast, and kidney function can push results either way.

Don't overread it

This was a single-centre research cohort with intermediate results excluded, so it does not give a real-world error rate.

The statistics, in plain English

An odds ratio of 1.63 per 10 mL/min means each 10-unit rise in eGFR raised the odds of a false negative by about 60%. The thyroid odds ratio (4.41, 1.36 to 14.3) has a very wide interval and did not survive age adjustment, so it is weaker evidence.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.