Blood biomarker standards move Alzheimer diagnosis closer to routine care
The clinical push for blood biomarker testing in Alzheimer disease now has a more concrete benchmark. As anti-amyloid therapies enter practice and biomarker confirmation becomes mandatory before treatment, the question is no longer whether blood tests are useful, but how accurate they must be to function safely in real-world care.
The Global CEO Initiative on Alzheimer’s Disease convened a workgroup to define minimum acceptable performance for blood biomarker tests of amyloid pathology. Amyloid PET served as the reference standard. The central recommendation is straightforward: a blood biomarker used as a triage test before confirmatory amyloid PET or cerebrospinal fluid testing should reach sensitivity of at least 90% and specificity of at least 85% in primary care, with specificity thresholds of 75% to 85% in secondary care depending on access to follow-up testing.
For a blood biomarker to stand alone as a confirmatory test, the bar rises to performance equivalent to CSF testing, with sensitivity and specificity of about 90%.
Why the diagnostic threshold matters now
In symptomatic Alzheimer disease, the arrival of disease-modifying therapy has changed the diagnostic sequence. A positive amyloid biomarker no longer serves only academic classification; it can determine whether a patient becomes eligible for treatment at a stage when intervention is most likely to matter. That makes misclassification expensive in every sense. False negatives can delay treatment until disease has progressed beyond the point where benefit is greatest. False positives can expose patients to unnecessary anxiety, further testing, or inappropriate treatment pathways.
Routine diagnosis of symptomatic Alzheimer disease already occurs under strain. Most diagnoses are made by clinicians who are not dementia specialists, and the diagnostic process often has to unfold in primary care or general neurology rather than in memory clinics. In that setting, a test that is too sensitive but not specific enough could overwhelm systems with confirmatory referrals. A test that is too specific but misses many true cases would fail to deliver the intended gain in access.
The workgroup’s performance targets reflect that balance. In primary care, high sensitivity protects against missed amyloid pathology, while high specificity limits unnecessary downstream workup. In secondary care, where access to confirmatory testing may differ, the acceptable specificity range broadens modestly, but the underlying principle remains the same: the blood test has to preserve diagnostic credibility before it can accelerate treatment access.
Amyloid pathology remains the anchor for clinical use
At present, clinical biomarker testing for symptomatic Alzheimer disease centers on amyloid pathology. Amyloid PET and CSF assays remain the principal tools used in routine care, while blood-based assays are increasingly entering the diagnostic conversation because they are more acceptable, more accessible, and easier to scale. The biological rationale is strong. Amyloid plaques are one of the defining neuropathological features of Alzheimer disease, and any test that tracks that pathology can help distinguish Alzheimer disease from other causes of cognitive impairment when the clinical picture is uncertain.
Still, a blood test is not interchangeable with a brain scan or lumbar puncture unless its accuracy is high enough to substitute for them in practice. Plasma measures vary widely in performance. Some assays may track amyloid biology closely; others may not. Without a minimum standard, broad uptake could bring uneven care, especially if a clinician assumes all commercially available tests are functionally equivalent.
The recommendations therefore act as a floor, not a ceiling. They are meant to define the minimum accuracy required for a test to be clinically useful, regardless of platform or manufacturer.
Predictive value depends on the patient in front of you
A single performance number can mislead if it is stripped from clinical context. Predictive values shift with pre-test probability, which means the same blood test can behave very differently in a patient with typical early Alzheimer symptoms than in someone with a mixed or atypical cognitive syndrome. A positive result carries more weight when the underlying probability of amyloid pathology is already high. A negative result is more reassuring when the pre-test probability is low.
This is why the workgroup emphasizes that biomarker results must be interpreted alongside the full clinical assessment. Cognitive symptoms often have multiple contributors, including vascular disease, medication effects, depression, sleep disorders, and other neurodegenerative syndromes. No Alzheimer biomarker test is intended to replace history, examination, cognitive testing, or laboratory evaluation for reversible causes of impairment.
The recommendation fits the clinical biology. Amyloid pathology is only one part of the disease process, even if it is the current biomarker anchor for diagnosis. A blood test that detects amyloid pathology can clarify one branch of the differential, but it cannot by itself explain every cognitive syndrome.
A framework designed for primary care and specialist settings
The workgroup based its recommendations on amyloid PET as the reference standard and examined how different levels of test performance would function across care settings. The resulting thresholds are meant to support use in both primary and secondary care, where the diagnostic workload is most likely to expand as anti-amyloid therapies become more common. This matters because most people with cognitive impairment are not evaluated by dementia subspecialists, and specialist capacity is already limited.
The practical implication is direct: blood biomarkers could help extend biomarker-based diagnosis into settings where amyloid PET and CSF testing are less available. That could reduce diagnostic delay, improve access to treatment, and help clinicians identify which patients truly need more invasive or resource-intensive confirmatory testing.
The biological logic is also clear. If blood biomarkers can reliably identify amyloid-positive patients early in the symptomatic phase, they can move the diagnosis closer to the point where disease-modifying therapy is still feasible. If they miss too many true cases, they lose that advantage immediately.
The new standard is about access as much as accuracy
The recommendations do not celebrate blood testing as a shortcut. They define a threshold for responsible use in a field where the consequences of error are now larger than they were before anti-amyloid treatment became clinically available. A test can be scalable, convenient, and still clinically inadequate if it cannot maintain the sensitivity and specificity needed for high-stakes decision-making.
The workgroup’s target product profile approach also signals what developers will need to show if blood biomarkers are to enter routine care in a durable way. They will need not only analytical validity but also clinically meaningful performance against a recognized standard. In a field where access is expanding faster than specialist supply, those thresholds may end up shaping which assays reach front-line practice and which remain research tools.
The remaining challenge is not whether blood biomarkers can be useful, but which assays can consistently meet the minimum standards across real-world patients, care settings, and pre-test probabilities. Until that is established, every blood result still has to be read in the context of the broader diagnostic workup.