Alzheimer’s Blood Test: What p-tau217 Measures and Who It Is Really For

The p-tau217 Alzheimer’s blood test is one of the most important diagnostic advances in recent years. In people with cognitive impairment, the best-validated assays can estimate brain amyloid pathology with accuracy approaching amyloid PET or cerebrospinal fluid biomarkers. That could shorten a diagnostic pathway that remains slow, expensive and unequal.

But a blood result does not turn a memory lapse into Alzheimer’s disease. On its own, it cannot tell us what is causing symptoms, how quickly they will progress or whether a healthy person will develop dementia. The useful question is not “Can I order this test?” but “What clinical decision would this result change for me?”

In May 2025, the FDA cleared the first blood-based device to aid diagnosis: the Lumipulse G pTau217/β-amyloid 1-42 plasma ratio. Its intended use is restricted to adults aged 55 or older with signs or symptoms of cognitive decline who are being assessed in specialist care. The FDA explicitly states that it is neither a screening test nor a stand-alone diagnosis.

Clinical and editorial update: July 29, 2026. This article is educational. Abnormal cognitive findings or biomarkers require interpretation by professionals experienced in neurology, geriatrics or memory care.

Quick answer: what p-tau217 can tell you

Question	Useful answer
Does it measure memory?	No. It measures a phosphorylated form of tau linked to Alzheimer biology. Memory requires clinical and cognitive assessment.
Does it detect amyloid plaques?	Indirectly. Validated assays are benchmarked against amyloid PET or cerebrospinal fluid biomarkers.
Does a positive result mean dementia?	No. Alzheimer pathology can exist without dementia, and symptomatic people may have more than one pathology.
Does a negative result end the workup?	Not always. High clinical suspicion, intermediate results or discordance may still require PET, lumbar puncture or other testing.
Is it for every healthy adult?	Not as routine screening. Individual prognostic use in asymptomatic people remains investigational.
Are all blood tests equivalent?	No. Analytes, platforms, calibration, cutoffs, sensitivity, specificity and validation populations differ.

What p-tau217 is and why it changed the field

Tau is a normal neuronal protein. In Alzheimer’s disease, its phosphorylation changes and abnormal tau tangles eventually accumulate inside cells. “p-tau217” means tau phosphorylated at the threonine 217 amino acid. Plasma levels rise in close relationship with the Alzheimer biological cascade, particularly when beta-amyloid is accumulating.

There is more than one way to measure it. Some assays quantify p-tau217 directly; others calculate its percentage relative to non-phosphorylated tau217; the FDA-cleared test uses a p-tau217-to-Aβ1-42 ratio. These numbers are not interchangeable. A cutoff from one platform should not be copied to another.

A 2024 Nature Medicine study in the BioFINDER-2 and Knight ADRC cohorts found that plasma percentage p-tau217 classified amyloid PET status with an area under the curve of 0.95-0.97. Among cognitively impaired participants, accuracy was approximately 89-90% and rose to 95% with a two-cutoff strategy that preserves an intermediate zone for confirmation.

The clinical leap is not only statistical. Until recently, biological confirmation relied mainly on amyloid PET — costly, scarce and involving radiation — or lumbar puncture. A scalable blood draw can bring first-line classification to many more people and reserve complex testing for those who actually need it.

What the FDA actually cleared

The 2025 clearance was based on 499 samples from cognitively impaired adults. Among people with a positive result, 91.7% had amyloid plaques confirmed by PET or cerebrospinal fluid. Among those with a negative result, 97.3% were also negative on the reference test. Fewer than 20% received an indeterminate result.

Those figures are impressive, but their population matters:

• adults aged 55 or older;
• with signs or symptoms of cognitive impairment;
• assessed in a specialist setting;
• with the result interpreted alongside clinical information;
• not as population screening or an isolated test.

Predictive value does not live inside the tube. It depends on the probability before testing. A positive result in a memory clinic, where objective impairment is common, means something different from the same result in a healthy adult who orders a wellness panel out of curiosity.

The 2025 guideline: triage, confirmation or neither

The first Alzheimer’s Association clinical practice guideline for blood biomarkers did not endorse a brand. It established performance thresholds for cognitively impaired patients in specialist memory care:

• Triage test: at least 90% sensitivity and 75% specificity. A negative can rule out pathology with high probability; a positive should be confirmed with amyloid PET or cerebrospinal fluid.
• Confirmatory test: at least 90% sensitivity and 90% specificity. In the appropriate context, it can substitute for amyloid PET or cerebrospinal fluid biomarkers.
• Good practice: never order the biomarker before a comprehensive clinical evaluation or interpret it without pretest probability.

The crucial warning is that many marketed products do not meet these thresholds. “Alzheimer’s blood test” can describe an automated, multicenter-validated assay or a commercial panel with uncertain individual utility. Naming the same biomarker does not guarantee the same performance.

What real-world primary and secondary care data show

A 2025 multicenter study evaluated automated p-tau217 in 1,767 people with cognitive symptoms across four specialist cohorts — including Barcelona — and a Swedish primary care cohort. Accuracy was 89-91% in secondary care and 85% in primary care. A two-cutoff approach raised accuracy to 92-94% while leaving roughly 12-17% of results in the intermediate zone.

A 2026 prospective study across seven memory clinics in Catalonia reinforced that trade-off. A single cutoff reached 84.9% accuracy. When researchers required very strict rule-out and rule-in thresholds, 42.7% of participants landed in the intermediate zone; the remaining 57.3% were classified with 95.3% accuracy. Greater certainty means more people need a second test.

The Spanish Society of Neurology’s 2026 consensus describes plasma biomarkers as especially promising because of access and diagnostic precision, while keeping systematic evaluation at the center. History, daily function, cognitive testing, examination, general laboratory work and structural imaging remain foundational.

The July 2026 study in symptom-free adults: important, but not a license to screen

The latest development appeared in JAMA on July 14, 2026. Researchers pooled 2,684 cognitively unimpaired older adults from six selected cohorts. Over a median 5.4 years, 478 progressed to cognitive impairment. Each one-standard-deviation increase in p-tau217 was associated with a 38% higher progression risk, and the association remained after accounting for amyloid measured by PET.

At five years, absolute risk was 12% in the low p-tau217 group, 15% in the intermediate group, 24% in the high group and 38% in the very-high group. Yet two-year risk was low in every category, at roughly 1-4%. Ten-year estimates were higher, but only 5% of participants had more than ten years of follow-up, so those numbers require particular caution.

This demonstrates prognostic value in research. It does not mean every symptom-free person should learn their p-tau217. Participants came from selected cohorts and trials, not a random population sample. We still need external validation, integrated models of age and vascular health, disclosure protocols, clearly indicated interventions and evidence that screening improves outcomes without creating psychological harm or diagnostic cascades.

Why a positive result may not fully explain symptoms

Alzheimer pathology becomes more common with age and can coexist with vascular disease, Lewy body disease, frontotemporal degeneration or other processes. Mixed pathology is common in older adults. A positive biomarker demonstrates a relevant biological process, but not necessarily the only cause.

There are also treatable or modifiable contributors to cognitive problems: sleep apnea, depression, anxiety, alcohol, sedating or anticholinergic medicines, hypothyroidism, vitamin B12 deficiency, hearing loss, pain, poor sleep, infection, epilepsy and vascular disease. A good assessment therefore does not begin with p-tau217. It begins by asking what changed, when, who noticed it and how daily life has been affected.

Perimenopausal brain fog, for example, can be highly disruptive without representing dementia. Likewise, finding microplastics in brain tissue does not let us assign causation in an individual. Responsible diagnosis avoids both dismissing symptoms and labeling them too early.

What a sensible diagnostic pathway looks like

1. Define the problem: episodic memory, language, orientation, attention, behavior, processing speed or executive function; onset and trajectory.
2. Establish objective impairment: validated cognitive testing and, where appropriate, neuropsychological assessment.
3. Look for alternatives and copathology: medicines, mood, sleep, hearing, laboratory markers, vascular risk and neurological examination.
4. Obtain structural imaging: MRI or CT may identify vascular disease, lesions, atrophy or other causes; neither is equivalent to amyloid PET.
5. Estimate pretest probability: age, clinical pattern, trajectory and history determine whether the biomarker can answer a useful question.
6. Select a validated assay: know the platform, sensitivity, specificity, intended population, thresholds and intermediate zone.
7. Confirm when necessary: use PET or cerebrospinal fluid for intermediate, discordant or treatment-critical decisions.
8. Disclose and decide: explain what the result proves, what it does not and what changes in follow-up, safety, treatment and planning.

p-tau217 compared with other biomarkers

Test	What it adds	Main limitation
Plasma p-tau217	High accuracy for Alzheimer pathology in validated assays.	Platforms and cutoffs are not interchangeable; it cannot explain the clinical picture alone.
Aβ42/Aβ40 or p-tau217/Aβ42	Estimates amyloid imbalance; some ratios reduce intermediate results.	Analytical and preanalytical variation differs by method.
GFAP	Astroglial response signal with supporting and research value.	Not specific to Alzheimer’s disease.
NfL	Reflects neuroaxonal injury and may support differential diagnosis or follow-up.	Rises in many neurological conditions and with age.
APOE	Genetic risk and safety/eligibility information for anti-amyloid therapy.	Does not diagnose Alzheimer’s and requires counseling about family implications.
PET or CSF	Reference tests for confirming amyloid/tau pathology.	Access, cost, PET radiation and lumbar puncture invasiveness.

What anti-amyloid treatment changes

Disease-modifying treatment makes biological confirmation more consequential. In the European Union, lecanemab is authorized for mild cognitive impairment or mild dementia due to Alzheimer’s in APOE ε4 non-carriers or people with one copy, with confirmed amyloid pathology. Donanemab also has a restricted indication and controlled-access safety measures. These are not preventive treatments for every adult and they carry meaningful risk.

A positive p-tau217 result does not automatically start treatment. Clinicians must establish clinical stage, pathology, APOE genotype, MRI findings, microhemorrhage burden, anticoagulant use and monitoring capacity. Amyloid-related imaging abnormalities — edema or hemorrhage known as ARIA — require rigorous selection and follow-up.

Questions to ask before paying for the test

• Do I have objective cognitive impairment or an unevaluated concern?
• Which assay and platform will the laboratory use?
• Has it been validated in people like me and in this clinical setting?
• What are its sensitivity, specificity and indeterminate-result rate?
• How do age, kidney function, comorbidities and pretest probability affect interpretation?
• Who will disclose the result and what support is available if it is positive?
• What changes after a low, intermediate or high result?
• Will I still need PET, lumbar puncture, MRI or APOE genotyping?

In a longevity clinic, adding more biomarkers does not automatically produce a better diagnosis. Quality means ordering the right test for the right person, integrating brain and vascular health and refusing to let one number replace medicine. Our guide to longevity biomarkers follows the same rule: a measurement earns its place when it changes a decision.

Conclusion: an entry point, not a verdict

p-tau217 can transform Alzheimer diagnosis by expanding access and reducing unnecessary PET scans and lumbar punctures. In people with objective cognitive impairment, within specialist assessment, high-performing assays now have genuine clinical utility.

Outside that context, caution remains part of the state of the art. A result is not a verdict, a positive is not dementia, and a biomarker detected years before symptoms does not yet provide a complete individual roadmap. The best brain medicine combines history, cognition, function, sleep, hearing, vascular risk, imaging and biomarkers around a specific question.

Sources

• U.S. Food and Drug Administration. FDA Clears First Blood Test Used in Diagnosing Alzheimer’s Disease. May 16, 2025. FDA.
• Palmqvist S, et al. Plasma phospho-tau217 for Alzheimer’s disease diagnosis in primary and secondary care using a fully automated platform. Nature Medicine. 2025;31:2036-2043. Study.
• Barthélemy NR, et al. Highly accurate blood test for Alzheimer’s disease is similar or superior to clinical cerebrospinal fluid tests. Nature Medicine. 2024;30:1085-1095. Study.
• Alzheimer’s Association. Clinical Practice Guideline on blood-based biomarkers in specialized care. Alzheimer’s & Dementia. 2025;21:e70535. Guideline.
• Buckley RF, et al. Prognostic Value of Blood-Based P-Tau217 Levels for Progression to Cognitive Impairment. JAMA. Published online July 14, 2026. Study.
• Spanish Society of Neurology. Consensus on the management of moderate-to-severe Alzheimer’s disease. 2026. Consensus.
• European Medicines Agency. Leqembi (lecanemab), current regulatory information. EMA.
• Spanish Agency of Medicines and Medical Devices. Therapeutic Positioning Report for Kisunla (donanemab). 2026. AEMPS.