Alzheimer's blood tests exist. In people with no symptoms, they answer the wrong question.

A blood test that can flag the biology of Alzheimer's disease is a genuine scientific advance, and it is now a real product. But for a person with no cognitive symptoms who wants to know whether they are "going to get Alzheimer's," these tests do not answer that question — and the bodies that vouch for them are explicit that they should not be used that way yet. The approved and recommended use is in people who already have symptoms, in specialist settings, with the result confirmed by another method.

What the test measures, and what it was cleared for

The first such test cleared by the U.S. Food and Drug Administration, Fujirebio's Lumipulse plasma ratio, was authorised on 16 May 2025 [s2]. Its cleared indication is narrow and worth reading precisely: it is an aid to identify amyloid pathology in adults aged 50 and older who are already presenting with symptoms of cognitive decline, in specialised care settings [s2]. It is not cleared as a screen for healthy adults, and it does not diagnose Alzheimer's disease on its own — it detects amyloid, one of the disease's biological hallmarks [s2].

That distinction — between detecting amyloid and diagnosing dementia — is the whole issue for an asymptomatic person.

Why a positive result means less than it sounds

Amyloid accumulates in the brain years, sometimes decades, before any symptoms, and a substantial share of cognitively normal older people have amyloid in their brains and never develop dementia in their lifetime. So in someone with no symptoms, a "positive" amyloid signal is not a diagnosis and not a prediction of certain decline. It is a risk marker — an elevated probability, over an uncertain timeframe, of a disease that may or may not arrive.

The statistics make this worse, not better. A test's positive predictive value — the chance that a positive result reflects true underlying pathology of the kind that matters — depends heavily on how common that pathology is in the group being tested. In a symptomatic memory-clinic population, where the pre-test probability is high, the test performs reasonably. Run the same test across the general asymptomatic population, where far fewer people have imminent disease, and the proportion of positives that are misleading rises sharply. A test can be accurate in the setting it was validated in and still generate a large share of frightening, low-value results when pointed at the worried well.

What the guidelines actually say

The Alzheimer's Association's appropriate-use recommendations for blood biomarkers are unusually direct about the limits [s1]. They recommend that these markers be used, cautiously, in specialised memory clinics as part of the diagnostic work-up of patients with cognitive symptoms, with results confirmed wherever possible by cerebrospinal fluid testing or PET imaging [s1]. They state that the markers should not yet be used as stand-alone diagnostic tests for Alzheimer's, and that additional data are needed before they can be considered for use in primary care [s1]. Outside of research, using them as pre-screeners is recommended only to identify people for trials — and even then, the recommendation is to confirm status with PET or CSF [s1].

Nowhere in that framework is there endorsement of testing an asymptomatic person to forecast their future. The guidance is built around people who already have symptoms and need a diagnosis explained.

The missing piece: nothing to do about it

The final reason the test answers the wrong question for a healthy person is that a positive result does not unlock a course of action that clearly changes the outcome. There is no proven treatment that prevents Alzheimer's in an amyloid-positive but cognitively normal person. The recently approved anti-amyloid antibody drugs are indicated for people with early symptomatic disease, not for asymptomatic amyloid carriers, and they carry their own significant risks. So a healthy person who tests positive is left with elevated anxiety, potential complications for insurance and long-term planning, and no evidence-based intervention that the result should trigger.

That calculus can shift. If a treatment is one day shown to help asymptomatic amyloid-positive people, the value of knowing changes with it, and large prevention trials are testing exactly that. Until then, the honest summary is that these tests are a real tool aimed at a specific job — helping diagnose people who already have cognitive symptoms — and that using them as a crystal ball for the symptom-free is a use their own developers and the relevant guidelines currently advise against [s1] [s2].

This article is informational and is not medical advice. Decisions about testing belong with a person and their clinician.