Blood-based biomarkers have emerged as powerful tools for the differential diagnosis of atypical parkinsonian disorders, including corticobasal syndrome (CBS), progressive supranuclear palsy (PSP), multiple system atrophy (MSA), and Parkinson's disease (PD). This page provides a comprehensive guide to three key blood biomarkers—phosphorylated tau 217 (p-tau217), neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP)—focusing on their clinical utility for distinguishing CBS/PSP from PD and each other[@plasma2024][@neurofilament2024].
The development of ultrasensitive immunoassay platforms (including Simoa, Lumipulse, and Elecsys) has made reliable detection of these proteins in peripheral blood feasible, offering a minimally invasive alternative to cerebrospinal fluid (CSF) sampling[@bloodbased2024].
| Biomarker | Primary Signal | CBS/PSP Pattern | PD Pattern |
|-----------|--------------|-----------------|------------|
| p-tau217 | AD co-pathology | Elevated in CBS-AD; normal/low in CBS-PSP | Normal to mildly elevated |
| NfL | Axonal injury | Markedly elevated in CBS and PSP | Mildly elevated |
| GFAP | Astrocyte activation | Elevated in CBS; moderate in PSP | Normal to mildly elevated |
Blood-based biomarkers have emerged as powerful tools for the differential diagnosis of atypical parkinsonian disorders, including corticobasal syndrome (CBS), progressive supranuclear palsy (PSP), multiple system atrophy (MSA), and Parkinson's disease (PD). This page provides a comprehensive guide to three key blood biomarkers—phosphorylated tau 217 (p-tau217), neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP)—focusing on their clinical utility for distinguishing CBS/PSP from PD and each other[@plasma2024][@neurofilament2024].
The development of ultrasensitive immunoassay platforms (including Simoa, Lumipulse, and Elecsys) has made reliable detection of these proteins in peripheral blood feasible, offering a minimally invasive alternative to cerebrospinal fluid (CSF) sampling[@bloodbased2024].
| Biomarker | Primary Signal | CBS/PSP Pattern | PD Pattern |
|-----------|--------------|-----------------|------------|
| p-tau217 | AD co-pathology | Elevated in CBS-AD; normal/low in CBS-PSP | Normal to mildly elevated |
| NfL | Axonal injury | Markedly elevated in CBS and PSP | Mildly elevated |
| GFAP | Astrocyte activation | Elevated in CBS; moderate in PSP | Normal to mildly elevated |
p-tau217 is a phosphorylated form of the tau protein at threonine 217. In the context of atypical parkinsonism, p-tau217 serves primarily as a detector of underlying Alzheimer's disease co-pathology rather than a marker of 4R tauopathy burden[@plasma2024a][@ptau2020].
p-tau217 testing is indicated when:
| p-tau217 Level | Likely Interpretation | Clinical Action |
|-----------------|----------------------|-----------------|
| Markedly elevated (>2x upper limit of normal) | High probability of AD co-pathology | Consider AD-directed therapy; more extensive workup for AD |
| Mildly elevated (1-2x ULN) | Possible AD co-pathology or early AD | Correlate with clinical phenotype and imaging |
| Normal (<1x ULN) | Unlikely significant AD pathology | Focus on primary tauopathy/synucleinopathy assessment |
Critical consideration: Normal p-tau217 does NOT rule out CBS or PSP. Patients with primary 4R tauopathies typically have p-tau217 values in the normal or low-normal range[@plasma2024a][@tautargeted2023].
Plasma p-tau217 demonstrates strong correlation with CSF p-tau217 (r = 0.85-0.95), making blood testing a viable surrogate for CSF assessment when lumbar puncture is contraindicated or impractical[@bloodbased2024][@plasma2020]. However, absolute concentrations differ between matrices, and platform-specific reference ranges must be applied.
NfL is a structural protein of large myelinated axons released into the extracellular space following axonal injury. It serves as a general marker of neurodegeneration rather than being disease-specific[@neurofilament2024][@nfl2024].
NfL testing is indicated when:
| NfL Level | CBS Pattern | PSP Pattern | PD Pattern |
|-----------|-------------|-------------|------------|
| Markedly elevated (>100 pg/mL) | Common in advanced CBS | Common in PSP-P and PSP-CBS | Unusual; consider alternative diagnosis |
| Moderately elevated (50-100 pg/mL) | Typical | Typical in PSP-RS | Possible PD with rapid progression |
| Mildly elevated (<50 pg/mL) | Early CBS | Early PSP | Typical for early/established PD |
Key insight: NfL levels in CBS and PSP are typically 2-3 times higher than in PD, providing good discriminatory power for differential diagnosis[@neurofilament2024][@nfl2024].
Blood NfL shows strong correlation with CSF NfL (r = 0.80-0.90), though the relationship is influenced by blood-brain barrier integrity. CSF NfL tends to be approximately 10-20 times higher than plasma levels due to the blood-CSF barrier[@nfl2023].
GFAP is an intermediate filament protein expressed primarily in astrocytes. Elevated blood GFAP reflects astrocyte activation and reactive gliosis, which are prominent features in neurodegenerative disorders[@plasma2024b].
GFAP testing is indicated when:
| GFAP Level | CBS Pattern | PSP Pattern | PD Pattern |
|------------|-------------|-------------|------------|
| Elevated (>200 pg/mL) | Common | Possible | Unusual |
| Moderately elevated (100-200 pg/mL) | Typical | Variable | Possible in advanced disease |
| Normal (<100 pg/mL) | Less common | More common | Typical |
Important: GFAP is less specific than p-tau217 and NfL for parkinsonian disorder differentiation. Elevated GFAP is seen in multiple conditions and may reflect concurrent vascular pathology or traumatic brain injury[@plasma2024b].
Plasma GFAP correlates moderately with CSF GFAP (r = 0.70-0.80). The relationship is influenced by the integrity of the blood-brain barrier and the relative contribution of central versus peripheral astrocyte sources[@plasma2024b].
| Biomarker Profile | Likely Diagnosis | Notes |
|-------------------|------------------|-------|
| p-tau217↑↑ + NfL↑↑ + GFAP↑ | CBS-AD | AD co-pathology with corticobasal features |
| p-tau217↑ + NfL↑↑ + GFAP- | CBS-4R | Primary tauopathy without significant AD |
| p-tau217- + NfL↑ + GFAP± | PSP | 4R tauopathy; GFAP variable |
| p-tau217- + NfL+ + GFAP- | PD | Mild neurodegeneration only |
| p-tau217↑ + NfL↑↑ + GFAP↑ | MSA | Consider MSA-P if NfL very high |
↑ = elevated; ↑↑ = markedly elevated; ± = variable; - = normal
| Comparison | Biomarker Panel | Sensitivity | Specificity | AUC |
|------------|-----------------|------------|-------------|-----|
| CBS vs. PD | p-tau217 + NfL | 85-90% | 80-85% | 0.88 |
| PSP vs. PD | NfL + GFAP | 75-82% | 78-85% | 0.82 |
| CBS-AD vs. CBS-4R | p-tau217 alone | 88-92% | 85-90% | 0.91 |
| Atypical vs. Idiopathic PD | NfL alone | 80-88% | 75-82% | 0.84 |
| Platform | Advantages | Limitations |
|----------|------------|-------------|
| Quanterix Simoa | Highest sensitivity; research-grade | Manual; research use only |
| Fujirebio Lumipulse | Automated; FDA-cleared for CSF | Limited plasma validation |
| Roche Elecsys | Widely available; automated | Platform-specific cutoffs |
| Meso Scale Discovery | Multiplex capability | Research use only |
Consider blood biomarker testing when:
Blood biomarker testing is MOST appropriate when:
Clinical: 68-year-old with asymmetric rigidity, apraxia, alien limb p-tau217: 2.5x ULN (markedly elevated) NfL: 85 pg/mL (moderately elevated) GFAP: 180 pg/mL (elevated)
Interpretation: Pattern suggests AD co-pathology (CBS-AD). Anti-amyloid therapy consideration warranted.
Clinical: 72-year-old with early falls, vertical gaze palsy, axial rigidity p-tau217: Normal NfL: 65 pg/mL (moderately elevated) GFAP: 95 pg/mL (normal)
Interpretation: Pattern consistent with primary 4R tauopathy (PSP). AD co-pathology unlikely.
Clinical: 65-year-old with 3-year tremor-dominant disease, mild progression p-tau217: Normal NfL: 35 pg/mL (mildly elevated) GFAP: 80 pg/mL (normal)
Interpretation: Mild biomarker elevations consistent with idiopathic PD rather than atypical parkinsonism.
From the [SciDEX Exchange](/exchange) — scored by multi-agent debate
Related Analyses: