What is tau PET, and how is it different from amyloid PET?

What it is

Tau PET - short for tau positron emission tomography - is a brain scan that uses an injected, FDA-approved imaging agent (a radiotracer) that binds to a protein called tau in the brain. The scan produces an image showing where in the brain tau has accumulated, and how much.

In Alzheimer's disease, tau is the second of the two proteins (alongside amyloid) that build up abnormally in the brain. Where amyloid tends to spread broadly across the cortex relatively early, tau accumulation typically follows a more orderly geographic pattern - moving from the medial temporal lobe (memory regions) outward to the broader cortex over time, in a sequence that has been characterised in pathological staging studies and is increasingly visible on PET imaging.

How it differs from amyloid PET

Amyloid PET, covered in our explainer on amyloid PET, tells the clinician whether the amyloid pathology associated with Alzheimer's is present at all. It is broadly a yes/no question with intensity gradations - a binary "is there Alzheimer's pathology underneath the symptoms" answer that reshapes the diagnostic picture.

Tau PET answers a different question. Once amyloid has confirmed that Alzheimer's pathology is present, tau PET can show how far the disease process has spread and where. That maps more directly to the clinical picture than amyloid does - the regions where tau has accumulated typically match the cognitive functions a patient is losing.

In practical terms:

• Amyloid PET is the better tool for "does this person have Alzheimer's pathology?"
• Tau PET is the better tool for "how advanced is the disease, where in the brain, and how does that match what we are seeing clinically?"

Both scans are PET, both involve a small radiotracer injection and a roughly hour-and-a-half visit, and both are interpreted by a radiologist or nuclear medicine specialist. The technical procedure is similar. The clinical question each one answers is different.

What the scan can and cannot tell you

It can tell you:

• The presence and pattern of tau accumulation across brain regions, in a way that is well-validated against post-mortem neuropathology.
• An indication of how advanced the underlying tau pathology is - quantified through standardised uptake values and regional patterns.
• Whether tau pathology is changing between two scans done at different points in time - used in clinical trials and increasingly in research settings.

It cannot tell you:

• Whether someone has Alzheimer's disease as a clinical condition independent of the rest of the picture. Tau pathology can also occur in other disorders, and the clinical reading still depends on putting the scan together with symptoms, history, and other tests.
• The exact cause of any specific symptom. As with all biomarker tools, the clinical picture has to be assembled, not read off a single image.
• The future trajectory of any individual patient with precision. Tau PET sharpens prognostic estimates, but it does not produce a fixed forecast.

When it is used

In contemporary US clinical practice, tau PET is most often used in three situations:

• Clinical-trial eligibility. Most contemporary tau-targeted Alzheimer's programs use tau PET to identify the right population - patients with amyloid confirmed and tau in a specified pattern or stage. We cover the tau pipeline in our Signal on the tau-targeting pipeline.
• Differentiating mixed pictures. When clinical symptoms and amyloid status do not fully line up, tau PET can clarify whether the tau distribution is consistent with Alzheimer's or with other tauopathies.
• Staging where the staging changes management. As more disease-modifying interventions become stage-sensitive, the question of how advanced the disease is can change which option is appropriate.

Outside these specific clinical and research contexts, tau PET is not yet in routine diagnostic workups in most settings - primarily because of access constraints rather than clinical limitations.

Why access to it is the next question

Tau PET tracer manufacturing, scanner availability, and reimbursement all sit further behind amyloid PET in the rollout curve. The result is that the test most useful for staging the disease is the test patients are least likely to be able to get. That gap is now an active issue in dementia coverage - both because clinical trials in the post-amyloid pipeline often require tau PET for enrolment, and because precision staging is becoming more clinically relevant. We track the access dynamics in our Signal on tau PET reimbursement as the next diagnostic access question.

If tau PET access opens up - through expanded reimbursement, broader tracer availability, or both - the diagnostic pathway becomes meaningfully more precise for a much larger fraction of patients. If it does not, the gap between what is technically possible and what is operationally available will continue to define who gets a precise diagnosis and who does not.

Coverage

US Medicare coverage of tau PET is currently more limited than coverage of amyloid PET, with status that has been evolving. Commercial coverage varies by plan. International coverage varies by country and is generally further behind US practice for this specific scan.

What this means in practice

If a clinician has recommended a tau PET scan, it is typically because either the staging information will change what is done next, or because clinical-trial eligibility depends on it. Knowing what the scan does and does not tell you helps patients and families read the result for what it is - a precise piece of the picture that complements the amyloid result, rather than replacing it.

For the broader diagnostic context, see our snapshot of diagnostic pathways for 2026. For the relationship between PET imaging and the cheaper, more accessible plasma biomarkers, see our explainer on plasma biomarkers.

This page is a plain-language primer. It is not medical advice. Decisions about diagnostic testing belong with the patient and their clinician.