Concept
Neurodegenerative Biomarkers
5 studiesEvidence last moved Sep 20, 2026
A neurodegenerative biomarker is a measurement used to detect, distinguish or track a neurodegenerative disease. The recurring problem in this cluster is that each measurement is a proxy whose relationship to the underlying pathology is itself under investigation.
Diagnosis in these diseases is being pushed earlier and toward measurement rather than clinical presentation, which raises the stakes on knowing what each marker actually binds to or reflects. Several of these studies are valuable precisely because they report where their own proxy fails.
Studies
5
Findings
4
4 supporting · 0 challenging · 0 qualifying citations
Open tensions
1
Latest change
Concept page published
Neurodegenerative Biomarkers
Currently
What we know
- Diagnostic separation and molecular specificity are different properties, and only the first was demonstrated.
- The behavioural null is what makes the electrophysiological difference interesting.
- The mild-disease figure is the one that matters for early diagnosis, and it is the weakest.
- The genotype effect existed only in one sex, and the cognitive outcomes were null in both.
Largest unresolved question
How close any of these markers is to clinical use differs by orders of magnitude, and the authors disagree with the framing more than with each other. Tau PET separates two diseases at 94.1% in a selected sample, while the Parkinson's model's mid-70s accuracy is explicitly not a standalone diagnostic test, and the glymphatic ALPS index is a diffusion proxy not shown to equal cerebrospinal fluid clearance.
Common misconceptions
A tau PET tracer measures tau.
18F-AV-1451 separated Alzheimer's disease from progressive supranuclear palsy at 94.1% accuracy, and post-mortem analysis complicated the assumption that its binding reflects tau specifically. Diagnostic usefulness does not establish what the tracer binds.
A lower ALPS index shows impaired glymphatic clearance.
The ALPS index is a diffusion measure along perivascular spaces. It differed by type 2 diabetes history, and the authors interpret that as impairment, but the study does not show that ALPS equals cerebrospinal fluid clearance, and the sample is small and cross-sectional.
Related
Claim ledger
What the evidence shows
Drawn from 5 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
Diagnostic separation and molecular specificity are different properties, and only the first was demonstrated.
A tau tracer separated two diseases with high accuracy while its binding target remained uncertain. 18F-AV-1451 PET distinguished Alzheimer's disease from progressive supranuclear palsy with 94.1% accuracy — cortical binding against midbrain and basal ganglia binding — yet post-mortem analysis complicated what the tracer binds.
The behavioural null is what makes the electrophysiological difference interesting.
A signal can be detectable before any symptom or behavioural difference. Asymptomatic C9orf72 carriers showed no group difference across six sustained-attention metrics, while during the N2 window they showed a larger negative potential at central sensors localising to bilateral precuneus and superior parietal regions.
The mild-disease figure is the one that matters for early diagnosis, and it is the weakest.
Machine learning on structural MRI reaches mid-range accuracy that improves with clinical data. MRI alone classified moderate-to-severe Parkinson's against controls at 74%, transfer learning raised mild Parkinson's against controls to 64%, and progression prediction exceeded 70% with MRI plus clinical features.
The genotype effect existed only in one sex, and the cognitive outcomes were null in both.
Risk genotype effects depend on sex in ways a main-effect analysis would miss. On a high-fat diet, male APOE4 mice were more susceptible to visceral fat accumulation and glucose intolerance than male APOE3 mice, while female APOE3 and APOE4 mice responded similarly — with no diet effects on anxiety-like behaviour or spatial memory in either genotype.
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
How close any of these markers is to clinical use differs by orders of magnitude, and the authors disagree with the framing more than with each other. Tau PET separates two diseases at 94.1% in a selected sample, while the Parkinson's model's mid-70s accuracy is explicitly not a standalone diagnostic test, and the glymphatic ALPS index is a diffusion proxy not shown to equal cerebrospinal fluid clearance.
How close any of these markers is to clinical use differs by orders of magnitude, and the authors disagree with the framing more than with each other. Tau PET separates two diseases at 94.1% in a selected sample, while the Parkinson's model's mid-70s accuracy is explicitly not a standalone diagnostic test, and the glymphatic ALPS index is a diffusion proxy not shown to equal cerebrospinal fluid clearance.
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
Concept page published
Neurodegenerative Biomarkers
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Rewordings are not listed.
- Concept page published
Papers
5 studies in this library bear on Neurodegenerative Biomarkers, ordered by citations.
- Can a new PET imaging tracer distinguish different tau diseases?
A new brain scanning chemical successfully maps and tells apart two different types of degenerative brain disorders based on where protein tangles build up.
- Does T2DM impair glymphatic flow on DTI-ALPS?
In 30 adults, longer T2DM history tracked with lower ALPS-index values, suggesting reduced perivascular water diffusivity.
- Does a high-fat diet hit APOE4 males harder?
In APOE knock-in mice, 12 weeks of 45% fat diet produced more visceral fat and glucose intolerance in male APOE4 than APOE3 mice, while females of both genotypes responded similarly and behavior did not change.
- Does C9orf72 already change parietal EEG before ALS symptoms?
In 87 asymptomatic relatives of C9orf72 ALS patients, carriers and non-carriers scored the same on a Go/NoGo attention task, but carriers showed stronger N2 negativity localized to precuneus and superior parietal cortex.
- Can a 3D MRI network spot and forecast Parkinson’s?
A 3D CNN using T1 MRI separated moderate-to-severe Parkinson’s disease from controls at 74% accuracy; transfer learning reached 64% for mild PD, and MRI plus clinical data predicted 2-year progression at >70%.
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
How close any of these markers is to clinical use differs by orders of magnitude, and the authors disagree with the framing more than with each other. Tau PET separates two diseases at 94.1% in a selected sample, while the Parkinson's model's mid-70s accuracy is explicitly not a standalone diagnostic test, and the glymphatic ALPS index is a diffusion proxy not shown to equal cerebrospinal fluid clearance.
Ask PaperFren about Neurodegenerative Biomarkers
Study this conceptflashcards and short-answer questions
Why is a behavioural null alongside a physiological difference more informative than either alone?
Because it locates the marker before the symptom. Asymptomatic C9orf72 carriers matched controls on all six sustained-attention metrics while showing a larger N2 negativity localising to precuneus and superior parietal cortex. Had behaviour also differed, the carriers would arguably be symptomatic. The null is what makes this a candidate preclinical marker — though cross-sectional EEG cannot show it predicts conversion.
A classifier reaches 74% for moderate-to-severe disease and 64% for mild disease. Which number matters?
The 64%, because early detection is the clinical problem. Moderate-to-severe Parkinson's is already diagnosable clinically, so a model that performs there adds little; the case for imaging-based diagnosis rests on identifying disease before it is obvious, and that is where performance drops closest to chance. The authors state the model is not a standalone diagnostic test.
What does a sex-dependent genotype effect imply for interpreting APOE4 risk?
That an unstratified estimate can describe neither sex. Male APOE4 mice showed more visceral fat accumulation and glucose intolerance than male APOE3 mice, while females showed no genotype difference — so a pooled effect would be an average of a real effect and none. The cognitive outcomes were null in both sexes, so this is a metabolic finding in mice over twelve weeks, not evidence about dementia risk in people.