Blood-Based Biomarkers in Dementia
The emergence of blood-based biomarkers represents one of the most transformative advances in dementia diagnosis. For decades, confirming Alzheimer disease (AD) pathology required either lumbar puncture for CSF analysis or amyloid PET—both expensive, invasive, or inaccessible in many settings. In May 2025, the FDA clearance of the Lumipulse G pTau217/Beta-Amyloid 1-42 Plasma Ratio (Fujirebio) marked the first regulatory approval of a blood test for AD, with commercial availability through Labcorp and Quest Diagnostics. Mass spectrometry–based tests such as PrecivityAD2 (C2N Diagnostics) have demonstrated 92% sensitivity and 96% specificity for detecting amyloid PET positivity. These advances, combined with the 2024 Alzheimer’s Association revised criteria that formally incorporate blood biomarkers as Core 1 diagnostic markers, are enabling a paradigm shift from specialist-dependent, imaging-reliant diagnosis to a scalable, blood-first screening approach deployable in primary care settings.
Bottom Line
- p-tau217 is the leading blood biomarker: Plasma phosphorylated tau 217 achieves an AUC of 0.93–0.96 for detecting AD pathology, outperforming p-tau181, p-tau231, and Aβ42/40 ratio as a standalone marker; the PrecivityAD2 test combining %p-tau217 with Aβ42/40 ratio achieves 92% sensitivity and 96% specificity
- FDA-cleared testing is now available: The Lumipulse G pTau217/Aβ1-42 Plasma Ratio was FDA-cleared in May 2025 and launched commercially through Labcorp and Quest Diagnostics in summer 2025
- Two-cutoff approach improves clinical utility: A low cutoff identifies likely amyloid-negative patients (NPV 83.6%) and a high cutoff identifies likely amyloid-positive patients (PPV 96.3%), with an intermediate zone (~20% of patients) requiring confirmatory PET or CSF testing
- GFAP and NfL provide complementary information: Plasma GFAP (astrocytic activation) becomes abnormal earlier than tau markers and may detect preclinical disease; plasma NfL (neurodegeneration) tracks disease severity and treatment response but is not AD-specific
- Pre-analytical factors are critical: Sample handling, collection tube material, renal function, and comorbidities can confound results; standardized protocols are essential for reliable interpretation
- Racial and ethnic variability exists: Significant differences in biomarker levels and predictive accuracy across racial and ethnic groups have been documented, and current cutoffs may not perform equally across all populations
- Clinical workflow integration: Blood biomarkers serve as a screening step before PET or CSF, reducing unnecessary invasive and costly testing; the 2024 AA revised criteria and the first clinical practice guideline (AAIC 2025) provide frameworks for implementation
Overview of Blood-Based Biomarkers
Blood-based biomarkers for AD can be organized according to the AT(N) framework expanded in the 2024 revised criteria. Each marker reflects a different aspect of disease biology, and their combined use provides a more comprehensive assessment than any single analyte alone.
| Biomarker | Category (ATN) | What It Measures | AUC for AD | AD Specificity | When It Changes |
|---|---|---|---|---|---|
| p-tau217 | T1 (tau phosphorylation) | Amyloid-driven tau phosphorylation at threonine 217 | 0.93–0.96 | High | Early, with amyloid accumulation |
| p-tau181 | T1 (tau phosphorylation) | Tau phosphorylation at threonine 181; reflects mature tangle pathology | 0.85–0.90 | Moderate–high | Early–mid stage |
| p-tau231 | T1 (tau phosphorylation) | Earliest-changing tau species; predominant in pre-tangles | ~0.85 | Moderate | Earliest tau marker (preclinical) |
| Aβ42/40 ratio | A (amyloid) | Reflects cerebral amyloid plaque burden (inversely correlated) | 0.82–0.87 | Moderate | Early (preclinical) |
| GFAP | I (inflammation) | Astrocyte activation and neuroinflammation | ~0.80–0.85 | Low (non-specific) | Earlier than tau markers |
| NfL | N (neurodegeneration) | Neuronal injury and axonal degeneration | ~0.75–0.80 | Low (non-specific) | With neurodegeneration |
Phosphorylated Tau 217 (p-tau217)
Plasma p-tau217 has emerged as the single most accurate blood biomarker for detecting AD pathology. Multiple head-to-head comparison studies have confirmed that mass spectrometry–based plasma p-tau217 outperforms all other plasma phospho-tau assays (including various platforms for p-tau181, p-tau217, and p-tau231) in predicting brain amyloid status and progression to AD dementia. Increased plasma p-tau217 has been shown to be equivalent to FDA-approved CSF biomarkers in predicting amyloid PET status.
Commercial Tests
| Feature | PrecivityAD2 (C2N Diagnostics) | Lumipulse G Plasma Ratio (Fujirebio) |
|---|---|---|
| Technology | Mass spectrometry | Fully automated immunoassay |
| Analytes reported | %p-tau217 and Aβ42/40 ratio (combined algorithm) | pTau217/Beta-Amyloid 1-42 plasma ratio |
| Performance | 92% sensitivity, 96% specificity for amyloid PET positivity | FDA-cleared cutoff: ratio ≤0.072 pg/mL for positivity (vs CSF Aβ42/40 reference) |
| Regulatory status | Laboratory-developed test (LDT) | FDA-cleared (May 16, 2025)—first FDA-cleared blood test for AD |
| Commercial availability | Available via C2N Diagnostics | Labcorp (August 2025); Quest Diagnostics (summer 2025) |
| Result format | Amyloid Probability Score (APS); reports likelihood of amyloid PET positivity | Ratio value with FDA-cleared positivity cutoff |
The Two-Cutoff Approach
Rather than a single binary threshold, emerging evidence supports a two-cutoff strategy for plasma p-tau217 that triages patients into three categories, analogous to the approach used in other laboratory diagnostics:
Two-Cutoff Triage Strategy for Plasma p-tau217
- Below low cutoff (0.229 pg/mL): Likely amyloid-negative; negative predictive value (NPV) of 83.6%; no further amyloid testing needed in most clinical contexts
- Above high cutoff (0.516 pg/mL): Likely amyloid-positive; positive predictive value (PPV) of 96.3%; may be sufficient for diagnosis depending on clinical scenario and treatment implications
- Intermediate zone: Approximately 20% of patients fall between cutoffs; confirmatory testing with amyloid PET or CSF biomarkers is recommended
- Overall accuracy: 91.1% across the full population using this two-cutoff strategy
- Clinical benefit: Reduces the need for confirmatory amyloid PET ($3,000–$6,000 per scan) or lumbar puncture by approximately 80% of cases, reserving these for the intermediate zone only
Phosphorylated Tau 181 (p-tau181)
Plasma p-tau181 was the first phospho-tau species widely studied in blood and remains an established biomarker for AD, reflecting mature neurofibrillary tangle pathology with an AUC of approximately 0.85–0.90. CSF p-tau181 is classified as a Core 1 biomarker in the 2024 revised criteria. Multiple commercial plasma assays are available, and the p-tau181/Aβ42 ratio improves discriminatory performance. However, head-to-head comparisons consistently demonstrate that p-tau217 outperforms p-tau181 in predicting both amyloid PET and tau PET status, making p-tau217 the preferred phospho-tau species for blood-based screening.
Phosphorylated Tau 231 (p-tau231)
Plasma p-tau231 is the earliest-changing tau biomarker, becoming abnormal at the lowest amyloid burden—before p-tau181 or p-tau217 reach significance. It predominates in pre-tangles (vs p-tau181 in mature neurofibrillary tangles) and predicts greater amyloid accumulation and cognitive decline over 3 years. Importantly, p-tau231 plateaus at moderate disease stages and does not track progression the way p-tau217 does, making it best suited for screening in preclinical stages rather than disease monitoring. It was included in the AAIC 2025 clinical practice guideline for blood biomarkers but is not yet FDA-cleared as a standalone test.
Amyloid-Beta 42/40 Ratio
The plasma Aβ42/40 ratio inversely correlates with brain amyloid burden—a lower ratio indicates greater cerebral amyloid deposition. The use of the ratio rather than absolute Aβ42 levels, combined with mass spectrometry measurement, has dramatically improved diagnostic accuracy compared with earlier ELISA-based approaches. In CSF, the difference in Aβ42 between AD patients and controls is approximately 50%, but in plasma it is only about 20%, amplifying the need for highly precise measurement. As a standalone marker, the plasma Aβ42/40 ratio achieves an AUC of approximately 0.82–0.87 for amyloid PET positivity. It performs best when combined with p-tau217 (as in PrecivityAD2 and the Lumipulse FDA-cleared test), and shows some of the earliest changes alongside p-tau231 in preclinical AD.
GFAP (Glial Fibrillary Acidic Protein)
Plasma GFAP is a marker of astrocyte activation and neuroinflammation corresponding to the “I” (inflammation) category of the expanded AT(N)ISV framework. It becomes abnormal earlier than tau biomarkers in the AD disease course, potentially identifying amyloid-related neuroinflammation before tau phosphorylation reaches detectable levels, and predicts future conversion from MCI to AD dementia.
Key Features of Plasma GFAP
- Measurement platform: Single-molecule array (Simoa) technology; not yet available on automated clinical immunoassay platforms
- Diagnostic accuracy: AUC ~0.80–0.85 for discriminating amyloid-positive from amyloid-negative individuals
- Temporal profile: Becomes abnormal earlier than p-tau181 and p-tau217 in the AD continuum, potentially marking the earliest astrocytic response to amyloid
- Specificity: Elevated in multiple neurodegenerative conditions (AD, FTD, vascular dementia, TBI) and is therefore not AD-specific
- Treatment response: Decreased plasma GFAP observed in both lecanemab and donanemab clinical trials compared with placebo, suggesting it may serve as a monitoring biomarker
- Regulatory status: Not FDA-cleared for AD diagnosis; primarily used in research settings
- Potential role: Early detection of amyloid-related neuroinflammation; disease progression monitoring; treatment response assessment
NfL (Neurofilament Light Chain)
Plasma neurofilament light chain (NfL) is a non-specific marker of neuronal injury and axonal degeneration corresponding to the “N” (neurodegeneration) category of the AT(N) framework. Elevated NfL correlates with the rate of cognitive decline, brain atrophy, and progression from MCI to dementia. Its primary value in AD lies in tracking disease severity rather than differential diagnosis, as NfL is elevated across a wide range of neurological conditions including FTD, ALS, MS, stroke, and traumatic brain injury. Notably, plasma NfL showed no significant treatment effect with either lecanemab or donanemab compared with placebo, which may reflect the fact that neuronal injury, once established, is less responsive to amyloid clearance than earlier pathologic processes.
Blood Biomarker Changes in Anti-Amyloid Therapy Trials
Clinical trials of anti-amyloid antibodies reveal which plasma biomarkers respond to treatment, informing their potential as monitoring biomarkers:
| Biomarker | Aducanumab | Lecanemab | Donanemab |
|---|---|---|---|
| Plasma Aβ42/40 ratio | Not reported | Increased | No treatment effect |
| Plasma p-tau181 | Decreased | Decreased | — |
| Plasma p-tau217 | — | — | Decreased |
| Plasma NfL | Not reported | No treatment effect | No treatment effect |
| Plasma GFAP | Not reported | Decreased | Decreased |
These data suggest that p-tau species and GFAP are the most promising candidates for monitoring therapeutic efficacy, while NfL may reflect irreversible neuronal damage less responsive to amyloid-directed therapy.
Pre-Analytical Considerations and Limitations
The clinical reliability of blood-based biomarkers depends critically on pre-analytical standardization. Blood specimens are subject to numerous variables that can influence results.
Pre-Analytical Factors That Affect Blood Biomarker Accuracy
- Collection tube material: Aβ peptides adhere to polystyrene (hard plastic) tubes, producing artifactually lower levels; polypropylene (soft plastic) collection tubes are required for reliable Aβ measurements, mirroring the same issue in CSF collection
- Sample processing time: Delays in centrifugation and plasma separation can alter biomarker levels; standardized protocols typically specify processing within 1–2 hours of collection
- Freeze-thaw cycles: Multiple freeze-thaw cycles degrade Aβ peptides and may alter p-tau measurements; samples should be aliquoted before freezing
- Renal function: Both p-tau181 and p-tau217 levels are artifactually elevated in patients with renal insufficiency, as demonstrated in the Mayo Clinic Study of Aging; estimated glomerular filtration rate (eGFR) should be considered when interpreting results
- Measurement platform: Mass spectrometry–based assays have demonstrated greater diagnostic accuracy than ELISA-based approaches, particularly for the Aβ42/40 ratio; platform-specific cutoffs are not interchangeable
- Fasting status and time of day: Emerging evidence suggests diurnal variation in some biomarkers, though standardized guidance is still evolving
- Comorbidities: Hepatic disease, diabetes, obesity, and cerebrovascular disease may influence biomarker levels independently of AD pathology
Racial and Ethnic Considerations
There are significant differences in the abundance and predictive value of fluid biomarkers across racial and ethnic groups. These differences must be considered when making diagnosis and treatment decisions for underrepresented groups.
Documented Racial and Ethnic Differences in Blood Biomarkers
- Plasma total tau: Cognitively unimpaired Mexican American individuals have higher levels than either African American or non-Hispanic White individuals
- Plasma Aβ42/40 ratio: Cognitively unimpaired African American individuals have a higher baseline ratio than non-Hispanic White individuals, potentially affecting the interpretation of cutoffs
- Plasma NfL: African American individuals with MCI show lower levels than non-Hispanic White individuals with MCI
- p-tau predictive accuracy: Plasma p-tau markers have been found to be less predictive of brain amyloid positivity among African American individuals compared with non-Hispanic White individuals
- CSF biomarkers: CSF total tau, p-tau181, and NfL are lower in African American participants, although Aβ42, Aβ40, and the Aβ42/40 ratio do not differ across groups
- Longitudinal data are encouraging: While baseline Aβ42/40 ratios differ, the longitudinal rate of change in plasma Aβ42/40 is similar across African American and non-Hispanic White individuals and is equally predictive of brain amyloid positivity over time
- Cutoff validation: Current FDA-cleared cutoffs have not been equally validated across all racial and ethnic populations; applying uniform thresholds may produce differential rates of false positives or negatives
When to Use Blood vs. CSF vs. PET
Blood-based biomarkers redefine the diagnostic workflow by enabling a tiered, cost-effective approach rather than eliminating the need for CSF or PET. The 2024 AA revised criteria and AAIC 2025 clinical practice guideline provide a framework for modality selection.
| Modality | Strengths | Limitations | Best Used For |
|---|---|---|---|
| Blood biomarkers | Non-invasive; scalable; low cost (~$200–$500); suitable for primary care settings; rapid turnaround | Intermediate zone requires confirmatory testing; affected by comorbidities; not yet validated for all populations; no biological staging capability | First-line screening; triage before PET or CSF; population-level screening in prevention trials |
| CSF biomarkers | Well-validated; ~90% concordance with amyloid PET; multiple analytes from single collection; can assess AD, CJD, and synucleinopathy simultaneously | Requires lumbar puncture; patient discomfort/refusal; not suitable for serial monitoring; collection tube requirements | Confirmatory testing after intermediate blood results; when multiple diagnoses need evaluation (AD + LBD + CJD); pre-treatment confirmation for anti-amyloid therapy |
| Amyloid PET | Direct visualization of plaque burden; quantifiable (Centiloids); FDA-approved for diagnosis and monitoring; spatial information | Expensive ($3,000–$6,000); limited availability; radiation exposure; no tau information; binary interpretation historically | Definitive amyloid confirmation; treatment monitoring (donanemab clearance assessment); intermediate blood/CSF results |
| Tau PET | Spatial distribution of tau pathology; enables biological staging; prognostic value; differentiates AD from non-AD tauopathies | Limited to specialized centers; expensive; single FDA-approved tracer (flortaucipir); not suitable for screening | Biological staging (Stages A–D); prognostication; increasing diagnostic confidence that AD contributes to symptoms |
Clinical Workflow Integration
The practical integration of blood-based biomarkers into clinical workflow follows a stepwise approach that maximizes diagnostic efficiency while minimizing unnecessary invasive testing:
- Clinical assessment: Cognitive complaint with objective testing (MoCA, MMSE); exclusion of reversible causes; structural brain MRI
- Blood biomarker screening: Plasma p-tau217 (± Aβ42/40 ratio) via FDA-cleared or validated LDT platform; apply two-cutoff interpretation
- Result-dependent triage: Below low cutoff → AD unlikely; above high cutoff → AD highly likely; intermediate zone → confirmatory amyloid PET or CSF
- Additional assessment: APOE genotyping (if anti-amyloid therapy considered); tau PET for biological staging; NfL for baseline neurodegeneration
- Treatment decision: Integrate biomarker results with clinical staging to determine disease-modifying therapy eligibility, ARIA risk, and MRI monitoring plan
This workflow represents a paradigm shift: the 2024 AA revised criteria formally recognize that a single abnormal plasma p-tau217 result can establish AD diagnosis. FDA-cleared plasma tests enable screening outside of specialized memory clinics, potentially reducing the 2–3 year average diagnostic delay. The first clinical practice guideline for blood biomarkers (AAIC 2025) provides standardized recommendations for test ordering and interpretation. Remaining barriers include variable insurance coverage, platform standardization, population-specific validation, and clinician education.
Direct-to-Consumer Testing
In July 2023, one manufacturer began offering blood-based biomarker testing directly to consumers, currently limited to the Aβ42/40 ratio. Consumers order the test online, undergo a blood draw at a local facility, and receive results through a patient portal. This development raises important practice considerations: patients may present with results they do not fully understand, and consumer test performance characteristics may differ from clinically validated assays. Clinician education on interpreting such results—particularly among diverse communities—will be essential as direct-to-consumer testing expands.
Limitations of Current Blood-Based Biomarkers
- Intermediate zone: ~20% of patients require confirmatory PET or CSF; high-stakes treatment decisions may still warrant confirmatory testing
- Comorbidity confounders: Renal insufficiency produces artifactually elevated p-tau181 and p-tau217; hepatic disease, diabetes, and obesity may also affect results
- Population-specific cutoffs not established: Validated primarily in non-Hispanic White cohorts; differences across racial/ethnic groups require population-specific validation
- Limited to AD pathology: No blood-based biomarker for alpha-synuclein (LBD), TDP-43 (FTD/LATE), or prion pathology (CJD) is clinically available
- Monitoring role not yet established: Validated cutoffs for monitoring therapeutic efficacy have not been defined for clinical practice
- Insurance coverage: Not yet universally covered by Medicare or commercial insurance
- Age-related amyloid positivity: Increases with age and does not always indicate symptomatic AD; clinical context is essential
References
- Quinn JF, Gray NE. Fluid biomarkers in dementia diagnosis. Continuum (Minneap Minn). 2024;30(6, Dementia):1790-1800.
- Jack CR Jr, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging of Alzheimer's disease: Alzheimer's Association Workgroup. Alzheimers Dement. 2024;20(8):5143-5169.
- Meyer MR, Kirmess KM, Eastwood S, et al. Clinical validation of the PrecivityAD2 blood test: a mass spectrometry-based test with algorithm combining p-tau217 and Abeta42/40 ratio. Alzheimers Dement. 2024;20(5):3179-3192.
- Barthelemy NR, Salvado G, Schindler SE, et al. Highly accurate blood test for Alzheimer's disease is similar or superior to clinical cerebrospinal fluid tests. Nat Med. 2024;30(4):1085-1095.
- Janelidze S, Bali D, Ashton NJ, et al. Head-to-head comparison of 10 plasma phospho-tau assays in prodromal Alzheimer's disease. Brain. 2023;146(4):1592-1601.
- Brum WS, Cullen NC, Janelidze S, et al. A two-step workflow based on plasma p-tau217 to screen for amyloid beta positivity with further confirmatory testing only in uncertain cases. Nat Aging. 2023;3(9):1079-1090.
- Mielke MM, Dage JL, Frank RD, et al. Performance of plasma phosphorylated tau 181 and 217 in the community. Nat Med. 2022;28(7):1398-1405.
- Brand AL, Lawler PE, Bollinger JG, et al. The performance of plasma amyloid beta measurements in identifying amyloid plaques in Alzheimer's disease: a literature review. Alzheimers Res Ther. 2022;14(1):195.
- Ashton NJ, Pascoal TA, Karikari TK, et al. Plasma p-tau231: a new biomarker for incipient Alzheimer's disease pathology. Acta Neuropathol. 2022;143(3):341-356.
- Benedet AL, Mila-Aloma M, Vrillon A, et al. Differences between plasma and cerebrospinal fluid glial fibrillary acidic protein levels across the Alzheimer disease continuum. JAMA Neurol. 2021;78(12):1471-1483.
- Cicognola C, Janelidze S, Hertze J, et al. Plasma glial fibrillary acidic protein detects Alzheimer pathology and predicts future conversion to Alzheimer dementia in patients with mild cognitive impairment. Alzheimers Res Ther. 2021;13(1):68.
- van Dyck CH, Swanson CJ, Aisen P, et al. Lecanemab in early Alzheimer's disease. N Engl J Med. 2023;388(1):9-21.
- Sims JR, Zimmer JA, Evans CD, et al. Donanemab in early symptomatic Alzheimer disease: the TRAILBLAZER-ALZ 2 randomized clinical trial. JAMA. 2023;330(6):512-527.
- Hall JR, Petersen M, Johnson L, O'Bryant SE. Characterizing plasma biomarkers of Alzheimer's in a diverse community-based cohort: a cross-sectional study of the HAB-HD cohort. Front Neurol. 2022;13:871947.
- Schindler SE, Karikari TK, Ashton NJ, et al. Effect of race on prediction of brain amyloidosis by plasma Abeta42/40, phosphorylated tau, and neurofilament light. Neurology. 2022;99(3):e245-e257.
- Xiong C, Schindler S, Luo J, et al. Baseline levels and longitudinal rates of change in plasma Abeta42/40 among self-identified Black/African American and White individuals. Res Sq. 2024;rs.3.rs-3783571.
- Mattsson-Carlgren N, Collij LE, Stomrud E, et al. Plasma biomarker strategy for selecting patients with Alzheimer disease for antiamyloid immunotherapies. JAMA Neurol. 2024;81(1):69-78.
- Arslan B, Zetterberg H, Ashton NJ. Blood-based biomarkers in Alzheimer's disease—moving towards a new era of diagnostics. Clin Chem Lab Med. 2024;62(6):1063-1069.
