Trend AnalysisMedicine & Health
Blood Biomarkers for Alzheimer's: p-tau217 and the End of Diagnostic Uncertainty
Diagnosing Alzheimer's disease (AD) has historically required either invasive lumbar puncture for cerebrospinal fluid (CSF) analysis or expensive PET imaging (~$3,000β5,000 per scan). Both are impract...
By Sean K.S. Shin
This blog summarizes research trends based on published paper abstracts. Specific numbers or findings may contain inaccuracies. For scholarly rigor, always consult the original papers cited in each post.
The Question
Diagnosing Alzheimer's disease (AD) has historically required either invasive lumbar puncture for cerebrospinal fluid (CSF) analysis or expensive PET imaging (~$3,000β5,000 per scan). Both are impractical for population-level screening. The 2024 NIA-AA revised criteria define AD biologically through core biomarkers β amyloid-beta (AΞ²) and tau β but applying these criteria at scale requires accessible, affordable tests. Plasma p-tau217 has emerged as the leading blood-based biomarker, with diagnostic accuracy approaching that of CSF and PET. Can a simple blood test transform Alzheimer's diagnosis from specialist-dependent to primary-care-accessible?
Landscape
Ghahremani et al. (2025) investigated the relationship between plasma p-tau217 and mild behavioural impairment (MBI) β late-onset neuropsychiatric symptoms that may precede cognitive decline. Their study found that elevated plasma p-tau217 is associated with MBI, suggesting that behavioural changes may serve as early clinical indicators of underlying AD pathology detectable by blood biomarkers. This work extends the clinical utility of p-tau217 beyond cognitively impaired populations to the preclinical stage.
Bandara et al. (2025) reviewed an alternative blood biomarker: glial fibrillary acidic protein (GFAP), a marker of astrocytic activation and neuroinflammation. While p-tau217 reflects tau pathology directly, GFAP captures a different pathological axis β the neuroinflammatory response that accompanies and may accelerate neurodegeneration. Their review focuses on GFAP as a standalone biomarker for AD neuropathology; the broader literature suggests combining GFAP with p-tau217 and AΞ²42/40 ratio may improve staging accuracy, though Bandara et al. do not specifically evaluate such multi-marker panels.
Wang et al. (2025) addressed the analytical challenge: detecting femtomolar concentrations of tau proteins in blood requires exquisitely sensitive assay platforms. Their review of electrochemical biosensors for blood-based tau detection documented emerging technologies that could enable point-of-care testing β moving AD screening from centralised reference laboratories to clinic-side devices.
Methods in Action
- Immunoassay platforms: Current clinical-grade p-tau217 assays (Lumipulse, Simoa, Elecsys) use antibody-based detection with chemiluminescence or single-molecule counting. Performance metrics include AUC >0.95 for distinguishing AΞ²-positive from AΞ²-negative individuals.
- Electrochemical biosensors: Wang et al.'s reviewed platforms use aptamer- or antibody-functionalised electrodes that generate measurable current changes upon tau binding. These offer potential for miniaturisation and point-of-care use.
- Cognitive-biomarker integration: Santos (2025) evaluated combining blood biomarkers with cognitive screening tools (MMSE, MoCA), showing that the combination improves diagnostic accuracy over either approach alone β particularly for distinguishing AD from other dementia subtypes.
- Novel biomarker candidates: Pollard (2025) introduced neuron-derived cell-free DNA (cfDNA) in blood plasma as a potential screening marker. When cortical neurons die, they release DNA fragments into the bloodstream. This approach could detect neurodegeneration regardless of underlying cause, complementing AD-specific biomarkers like p-tau217.
Key Claims & Evidence
<
| Claim | Evidence | Verdict |
|---|
| Plasma p-tau217 has AUC >0.95 for amyloid positivity | Multi-cohort validation studies across diverse populations | Confirmed; performance approaches CSF and PET |
| p-tau217 detects preclinical AD | Association with MBI in cognitively normal individuals (Ghahremani et al. 2025) | Supported; longitudinal validation ongoing |
| Multi-marker panels outperform single biomarkers | p-tau217 + GFAP + AΞ²42/40 improves staging accuracy (broader literature) | Supported; Bandara et al. 2025 focuses on GFAP standalone; optimal panel composition under active investigation |
| Point-of-care testing is feasible | Electrochemical biosensors achieve relevant sensitivity (Wang et al. 2025) | Technologically promising; clinical validation pending |
| Neuron-derived cfDNA is a viable AD biomarker | Preliminary data showing correlation with neurodegeneration (Pollard 2025) | Very early stage; requires large cohort validation |
Open Questions
Cut-off standardisation: Different assay platforms use different antibodies and calibrators, producing different absolute p-tau217 values. Can universal cut-offs be established, or will platform-specific thresholds persist?
Population diversity: Most validation cohorts are predominantly white and highly educated. How does p-tau217 perform across ethnic groups, socioeconomic backgrounds, and comorbidity profiles?
Clinical utility: Detecting preclinical AD is valuable only if it leads to actionable interventions. With anti-amyloid therapies (lecanemab, donanemab) now available but modestly effective, does early diagnosis improve outcomes or create anxiety without benefit?
Health system readiness: If blood-based AD screening becomes routine, healthcare systems will face millions of new diagnoses requiring specialist follow-up. Are neurology and geriatric services prepared for this volume?What This Means for Your Research
For neurologists and geriatricians, plasma p-tau217 is approaching clinical readiness for triaging patients into those who need (vs. do not need) confirmatory PET or CSF testing β a "rule-out" function that could reduce unnecessary invasive procedures by 30β50%. For diagnostics researchers, the multi-marker panel approach (combining tau, amyloid, neuroinflammation, and neurodegeneration markers) is where the highest-impact work lies. For public health researchers, the equity implications of blood-based AD testing β who gets tested, who pays, and what happens after a positive result β are urgent questions as these tests move toward regulatory approval.
Referenced Papers
- [1] Wang, J. et al. (2025). Electrochemical Technology for the Detection of Tau Proteins as a Biomarker of Alzheimer's Disease in Blood. Biosensors, 15(2), 85. DOI: 10.3390/bios15020085
- [2] Bandara, E. et al. (2025). The Role of GFAP in AD Neuropathology and Its Potential as a Blood Biomarker. Molecular Neurobiology. DOI: 10.1007/s12035-025-05219-3
- [3] Ghahremani, M. et al. (2025). Exploring the association between mild behavioral impairment and plasma pβtau217: Implications for early detection of Alzheimer's disease. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring. DOI: 10.1002/dad2.70119
- [4] Santos, M. (2025). Early Detection of Alzheimer's Disease: Biomarkers and Cognitive Screening Tools. DOI: 10.58840/xaa0wd39
- [5] Pollard, C.A. (2025). Detection of neuron-derived cfDNA in blood plasma: A novel screening approach for AD. Alzheimer's & Dementia (conference abstract). DOI: 10.1002/alz70856_103924
λ©΄μ±
μ‘°ν: μ΄ κ²μλ¬Όμ μ 보 μ 곡 λͺ©μ μ μ°κ΅¬ λν₯ κ°μμ΄λ€. μΈμ©λ νΉμ μ°κ΅¬ κ²°κ³Ό, ν΅κ³, μ£Όμ₯μ νμ μ°κ΅¬μμ μΈμ©νκΈ° μ μ μλ³Έ λ
Όλ¬Έκ³Ό λμ‘°νμ¬ κ²μ¦ν΄μΌ νλ€.
μμΈ νμ΄λ¨Έ νμ‘ λ°μ΄μ€λ§μ»€: p-tau217κ³Ό μ§λ¨ λΆνμ€μ±μ μ’
μ
λΆμΌ: μν | λ°©λ²λ‘ : μμ-μ§λ¨
μ μ: Sean K.S. Shin | λ μ§: 2026-03-17
μ°κ΅¬ μ§λ¬Έ
μμΈ νμ΄λ¨Έλ³(AD) μ§λ¨μ μμ¬μ μΌλ‘ λμ²μμ‘(CSF) λΆμμ μν μΉ¨μ΅μ μμΆ μ²μ λλ κ³ κ°μ PET μμ(μ€μΊλΉ μ½ $3,000β5,000)μ νμλ‘ νλ€. λ λ°©λ² λͺ¨λ μΈκ΅¬ μμ€μ μ λ³ κ²μ¬μλ μ€μ©μ μ΄μ§ μλ€. 2024λ
NIA-AA κ°μ κΈ°μ€μ ν΅μ¬ λ°μ΄μ€λ§μ»€μΈ μλ°λ‘μ΄λ-λ² ν(AΞ²)μ νμ°λ₯Ό ν΅ν΄ ADλ₯Ό μλ¬Όνμ μΌλ‘ μ μνμ§λ§, μ΄ κΈ°μ€μ λκ·λͺ¨λ‘ μ μ©νλ €λ©΄ μ κ·Ό κ°λ₯νκ³ μ λ ΄ν κ²μ¬κ° νμνλ€. νμ₯ p-tau217μ CSF λ° PETμ κ·Όμ νλ μ§λ¨ μ νλλ₯Ό 보μ΄λ©° μ λμ μΈ νμ‘ κΈ°λ° λ°μ΄μ€λ§μ»€λ‘ λΆμνλ€. λ¨μν νμ‘ κ²μ¬κ° μμΈ νμ΄λ¨Έ μ§λ¨μ μ λ¬Έμ μμ‘΄μμ μΌμ°¨ μ§λ£ μ κ·Ό κ°λ₯ν λ°©μμΌλ‘ μ νν μ μμκΉ?
μ°κ΅¬ νν©
Ghahremani λ±(2025)μ νμ₯ p-tau217κ³Ό κ²½λ νλ μ₯μ (MBI) β μΈμ§ μ νμ μ νν μ μλ νκΈ° λ°λ³ μ κ²½μ μ κ³Όμ μ¦μ β μ¬μ΄μ κ΄κ³λ₯Ό μ‘°μ¬νλ€. μ΄ μ°κ΅¬λ νμ₯ p-tau217 μμΉμ΄ MBIμ μ°κ΄λμ΄ μμμ λ°κ²¬νμΌλ©°, μ΄λ νλ λ³νκ° νμ‘ λ°μ΄μ€λ§μ»€λ‘ κ°μ§ κ°λ₯ν κΈ°μ AD λ³λ¦¬μ μ΄κΈ° μμ μ§ν μν μ ν μ μμμ μμ¬νλ€. μ΄ μ°κ΅¬λ p-tau217μ μμμ μ μ©μ±μ μΈμ§ μ₯μ μ§λ¨μ λμ΄ μ μμ λ¨κ³κΉμ§ νμ₯νλ€.
Bandara λ±(2025)μ λμμ νμ‘ λ°μ΄μ€λ§μ»€μΈ μ κ²½κ΅ μ¬μ μ§ μ°μ± λ¨λ°±μ§(GFAP)μ κ²ν νλ€. GFAPλ μ±μμΈν¬ νμ±ν λ° μ κ²½μΌμ¦μ λ§μ»€μ΄λ€. p-tau217μ΄ νμ° λ³λ¦¬λ₯Ό μ§μ λ°μνλ λ°λ©΄, GFAPλ μ κ²½ν΄νμ μλ°νκ³ κ°μνν μ μλ μ κ²½μΌμ¦ λ°μμ΄λΌλ λ€λ₯Έ λ³λ¦¬ μΆμ ν¬μ°©νλ€. μ΄λ€μ κ²ν λ AD μ κ²½λ³λ¦¬μ λν λ
립μ λ°μ΄μ€λ§μ»€λ‘μμ GFAPμ μ΄μ μ λ§μΆκ³ μμΌλ©°, λ λμ λ¬Ένμμλ GFAPλ₯Ό p-tau217 λ° AΞ²42/40 λΉμ¨κ³Ό κ²°ν©νλ©΄ λ³κΈ° κ²°μ μ νλκ° ν₯μλ μ μμμ μμ¬νμ§λ§, Bandara λ±μ μ΄λ¬ν λ€μ€ λ§μ»€ ν¨λμ ꡬ체μ μΌλ‘ νκ°νμ§λ μλλ€.
Wang λ±(2025)μ λΆμμ κ³Όμ λ₯Ό λ€λ£¨μλ€. νμ‘ λ΄ ν¨ν λͺ° λλμ νμ° λ¨λ°±μ§μ κ²μΆνλ €λ©΄ κ·Ήλλ‘ λ―Όκ°ν λΆμ νλ«νΌμ΄ νμνλ€. μ΄λ€μ νμ‘ κΈ°λ° νμ° κ²μΆμ μν μ κΈ°ννμ λ°μ΄μ€μΌμ κ²ν λ AD μ λ³ κ²μ¬λ₯Ό μ€μνλ κΈ°μ€ μ€νμ€μμ μ§λ£μ νμ₯ κΈ°κΈ°λ‘ μ΄λμν¬ μ μλ μ ν₯ κΈ°μ λ€μ κΈ°λ‘νλ€.
λ°©λ²λ‘ μ μ μ©
- λ©΄μλΆμ νλ«νΌ: νμ¬ μμ λ±κΈμ p-tau217 λΆμ(Lumipulse, Simoa, Elecsys)μ ννλ°κ΄ λλ λ¨μΌ λΆμ κ³μλ₯Ό μ΄μ©ν ν체 κΈ°λ° κ²μΆμ μ¬μ©νλ€. μ±λ₯ μ§νμλ AΞ² μμ± κ°μΈκ³Ό AΞ² μμ± κ°μΈμ ꡬλ³νλ AUC >0.95κ° ν¬ν¨λλ€.
- μ κΈ°ννμ λ°μ΄μ€μΌμ: Wang λ±μ΄ κ²ν ν νλ«νΌμ νμ° κ²°ν© μ μΈ‘μ κ°λ₯ν μ λ₯ λ³νλ₯Ό μμ±νλ μ±νλ¨Έ λλ ν체 κΈ°λ₯ν μ κ·Ήμ μ¬μ©νλ€. μ΄λ μνν λ° νμ₯ μ§λ£ μ¬μ©μ κ°λ₯μ±μ μ 곡νλ€.
- μΈμ§-λ°μ΄μ€λ§μ»€ ν΅ν©: Santos(2025)λ νμ‘ λ°μ΄μ€λ§μ»€μ μΈμ§ μ λ³ λꡬ(MMSE, MoCA)μ κ²°ν©μ νκ°νμ¬, νΉν ADλ₯Ό λ€λ₯Έ μΉλ§€ μνκ³Ό ꡬλ³νλ λ° μμ΄ λ μ κ·Όλ²μ λ¨λ
μΌλ‘ μ¬μ©ν λλ³΄λ€ κ²°ν© μ μ§λ¨ μ νλκ° ν₯μλ¨μ 보μλ€.
- μλ‘μ΄ λ°μ΄μ€λ§μ»€ ν보: Pollard(2025)λ νμ₯ λ΄ λ΄λ° μ λ 무μΈν¬ DNA(cfDNA)λ₯Ό μ μ¬μ μ λ³ λ§μ»€λ‘ μ μνλ€. νΌμ§ λ΄λ°μ΄ μ¬λ©Ένλ©΄ νλ₯λ‘ DNA λ¨νΈμ λ°©μΆνλ€. μ΄ μ κ·Όλ²μ κ·Όλ³Έμ μΈ μμΈμ κ΄κ³μμ΄ μ κ²½ν΄νμ κ°μ§ν μ μμ΄ p-tau217κ³Ό κ°μ AD νΉμ΄μ λ°μ΄μ€λ§μ»€λ₯Ό 보μν μ μλ€.
μ£Όμ μ£Όμ₯ λ° κ·Όκ±°
<
| μ£Όμ₯ | κ·Όκ±° | νμ |
|---|
| νμ₯ p-tau217μ μλ°λ‘μ΄λ μμ±μ λν΄ AUC >0.95λ₯Ό λνλΈλ€ | λ€μν μ§λ¨μ λμμΌλ‘ ν λ€μ€ μ½νΈνΈ κ²μ¦ μ°κ΅¬ | νμΈλ¨; μ±λ₯μ CSF λ° PETμ κ·Όμ νλ€ |
| p-tau217μ μ μμ ADλ₯Ό κ°μ§νλ€ | μΈμ§μ μΌλ‘ μ μμΈ κ°μΈμμ MBIμμ μ°κ΄μ± (Ghahremani et al. 2025) | μ§μ§λ¨; μ’
λ¨ κ²μ¦ μ§ν μ€ |
| λ€μ€ λ§μ»€ ν¨λμ΄ λ¨μΌ λ°μ΄μ€λ§μ»€λ³΄λ€ μ°μνλ€ | p-tau217 + GFAP + AΞ²42/40μ΄ λ³κΈ° μ νλλ₯Ό ν₯μμν¨λ€ (κ΄λ²μν λ¬Έν) | μ§μ§λ¨; Bandara et al. 2025λ GFAP λ¨λ
μ μ§μ€νλ©°, μ΅μ ν¨λ ꡬμ±μ νλ°ν μ°κ΅¬ μ€μ΄λ€ |
| νμ₯μ§λ£ κ²μ¬(Point-of-care testing)κ° μ€ν κ°λ₯νλ€ | μ κΈ°νν λ°μ΄μ€μΌμκ° κ΄λ ¨ λ―Όκ°λλ₯Ό λ¬μ±νλ€ (Wang et al. 2025) | κΈ°μ μ μΌλ‘ μ λ§νλ; μμ κ²μ¦μ΄ λκΈ° μ€μ΄λ€ |
| λ΄λ° μ λ cfDNAλ μ€ν κ°λ₯ν AD λ°μ΄μ€λ§μ»€μ΄λ€ | μ κ²½ν΄νκ³Όμ μκ΄κ΄κ³λ₯Ό 보μ΄λ μλΉ λ°μ΄ν° (Pollard 2025) | λ§€μ° μ΄κΈ° λ¨κ³μ΄λ©°; λκ·λͺ¨ μ½νΈνΈ κ²μ¦μ΄ νμνλ€ |
λ―Έν΄κ²° κ³Όμ
μ»·μ€ν νμ€ν: μλ‘ λ€λ₯Έ λΆμ νλ«νΌμ΄ μλ‘ λ€λ₯Έ ν체μ κ΅μ λ¬Όμ§μ μ¬μ©νμ¬ μλ‘ λ€λ₯Έ μ λμ p-tau217 κ°μ μμ±νλ€. λ²μ© μ»·μ€νλ₯Ό ν립ν μ μλκ°, μλλ©΄ νλ«νΌλ³ μκ³κ°μ΄ μ§μλ κ²μΈκ°?
μ§λ¨ λ€μμ±: λλΆλΆμ κ²μ¦ μ½νΈνΈλ μ£Όλ‘ λ°±μΈμ΄λ©° κ³ νλ ₯μλ‘ κ΅¬μ±λμ΄ μλ€. p-tau217μ λ―Όμ‘± μ§λ¨, μ¬νκ²½μ μ λ°°κ²½, λλ° μ§ν νλ‘νμΌμ λ°λΌ μ΄λ ν μ±λ₯μ 보μ΄λκ°?
μμμ μ μ©μ±: μ μμ ADλ₯Ό κ°μ§νλ κ²μ μ€ν κ°λ₯ν μ€μ¬λ‘ μ΄μ΄μ§ κ²½μ°μλ§ κ°μΉκ° μλ€. νμλ°λ‘μ΄λ μΉλ£μ (lecanemab, donanemab)κ° νμ¬ μ΄μ© κ°λ₯νμ§λ§ ν¨κ³Όκ° μ νμ μΈ μν©μμ, μ‘°κΈ° μ§λ¨μ΄ κ²°κ³Όλ₯Ό κ°μ νλκ°, μλλ©΄ ν¨μ΅ μμ΄ λΆμλ§μ μΌκΈ°νλκ°?
μλ£ μμ€ν
μ€λΉλ: νμ‘ κΈ°λ° AD μ λ³ κ²μ¬κ° μΌμνλ κ²½μ°, μλ£ μμ€ν
μ μ λ¬Έμ νμ μ§λ£κ° νμν μλ°±λ§ κ±΄μ μλ‘μ΄ μ§λ¨μ μ§λ©΄νκ² λ κ²μ΄λ€. μ κ²½κ³Ό λ° λ
ΈμΈμν μλΉμ€κ° μ΄ κ·λͺ¨μ λλΉλμ΄ μλκ°?μ°κ΅¬μμκ² μ£Όλ μλ―Έ
μ κ²½κ³Ό μ λ¬Έμμ λ
ΈμΈμν μ λ¬Έμμκ² μμ΄, νμ₯ p-tau217μ νμ¦μ PET λλ CSF κ²μ¬κ° νμν(λλ νμνμ§ μμ) νμλ₯Ό μ λ³νλ μμ μ€λΉ λ¨κ³μ κ·Όμ νκ³ μμΌλ©°, μ΄λ λΆνμν μΉ¨μ΅μ μμ μ 30~50% κ°μμν¬ μ μλ "λ°°μ " κΈ°λ₯μ νλ€. μ§λ¨ μ°κ΅¬μμκ²λ νμ°, μλ°λ‘μ΄λ, μ κ²½μΌμ¦, μ κ²½ν΄ν λ§μ»€λ₯Ό κ²°ν©ν λ€μ€ λ§μ»€ ν¨λ μ κ·Όλ²μ΄ κ°μ₯ λμ μν₯λ ₯μ μ§λ μ°κ΅¬ λΆμΌμ΄λ€. κ³΅μ€ λ³΄κ±΄ μ°κ΅¬μμκ²λ νμ‘ κΈ°λ° AD κ²μ¬μ ννμ± λ¬Έμ , μ¦ λκ° κ²μ¬λ₯Ό λ°λμ§, λκ° λΉμ©μ λΆλ΄νλμ§, μμ± κ²°κ³Ό μ΄νμλ μ΄λ ν μΌμ΄ λ°μνλμ§μ λν μ§λ¬Έμ΄ μ΄ κ²μ¬λ€μ΄ κ·μ μΉμΈμ ν₯ν΄ λμκ°λ©΄μ μκΈν κ³Όμ λ‘ λΆμνκ³ μλ€.
References (5)
Wang, J., Lu, X., & He, Y. (2025). Electrochemical Technology for the Detection of Tau Proteins as a Biomarker of Alzheimerβs Disease in Blood. Biosensors, 15(2), 85.
Bandara, E. M. S., Asih, P. R., Pedrini, S., Hone, E., Fernando, W. M. A. D. B., & Martins, R. N. (2025). The Role of Glial Fibrillary Acidic Protein in the Neuropathology of Alzheimerβs Disease and Its Potential as a Blood Biomarker for Early Diagnosis and Progression. Molecular Neurobiology, 62(12), 15576-15608.
Ghahremani, M., Leon, R., Smith, E. E., & Ismail, Z. (2025). Exploring the association between mild behavioral impairment and plasma pβtau217: Implications for early detection of Alzheimer's disease. Alzheimer's & Dementia: Diagnosis, Assessment & Disease Monitoring, 17(2).
Santos, M. (2025). Early Detection of Alzheimerβs Disease: Biomarkers and Cognitive Screening Tools. OTS Canadian Journal, 4(6), 74.
Pollard, C. (2025). Detection of neuronβderived cfDNA in blood plasma: A novel screening and monitoring approach for Alzheimer's disease and other neurodegenerative conditions. Alzheimer's & Dementia, 21(S2).