Utilizing Intraindividual Cognitive Variability to Predict Early Neuronal Synuclein Disease Progression

This study demonstrates that intraindividual cognitive variability, specifically measured by the coefficient of variation, serves as a sensitive marker that distinguishes prodromal neuronal synuclein disease from healthy controls and independently predicts short-term progression to more advanced stages.

The Gist

The Big Picture: Catching the "Ghost" Before It Haunts the House

Imagine Parkinson's disease (and related conditions) not as a sudden storm, but as a slow-moving fog. Long before the fog gets thick enough to stop you from walking or talking, it starts to obscure your vision slightly. Scientists call this the "prodromal" or pre-symptomatic stage.

The goal of this study was to find a way to see that fog before it becomes a storm. The researchers looked at a specific clue: how inconsistent a person's brain is.

The Main Character: "Intraindividual Variability" (IIV)

Let's call this concept "The Drunk Driver Test" (but for your brain).

Imagine you are taking a driving test.

• The Healthy Brain: You drive at a steady 60 mph the whole time. Sometimes you hit 61, sometimes 59, but you are very consistent.
• The Early Disease Brain: You are driving 60 mph, then suddenly 40, then 75, then 50. Your average speed might still look okay (maybe 58 mph), but your inconsistency is the problem. You are wobbling.

In this study, the researchers call this wobbling Intraindividual Variability (IIV). It's not about how smart you are on average; it's about how unreliable your brain is from one moment to the next.

The Experiment: The "Cognitive Obstacle Course"

The researchers used data from a massive study called PPMI (Parkinson's Progression Markers Initiative). They looked at nearly 1,000 people:

1. Healthy People: No signs of Parkinson's.
2. Stage 2 Patients: People who have the biological markers of Parkinson's (like a specific protein in their spinal fluid) and tiny symptoms, but they can still function normally in daily life. They are in the "foggy" stage.

They gave everyone a battery of 11 different mental tests (memory, math, word finding, etc.). Instead of just looking at the final score, they looked at the spread of the scores.

The Analogy:
Imagine a student taking 11 math quizzes.

• Student A (Healthy): Gets 95, 96, 94, 95, 95... Very consistent.
• Student B (Early Parkinson's): Gets 95, 60, 98, 40, 92... Their average might be 77, which looks like a "C" student. But the real story is the chaos. They are brilliant one minute and blank the next.

What They Found

1. The Wobble Starts Early: Even though the "Stage 2" patients looked normal on average, their brains were much more "wobbly" (inconsistent) than the healthy people. They were the ones driving 60, then 40, then 75.
2. The Wobble Predicts the Future: The researchers waited one year to see who got worse.
   • The people who stayed stable had steady brains.
   • The people who progressed to a more advanced stage (Stage 3+) were the ones with the most wobbly brains at the start.
3. It's a Better Crystal Ball Than Average Scores: If you only looked at the average test score, you might miss these people. But if you looked at the variability (the inconsistency), you could predict who was going to get sick faster.

Why This Matters (The "So What?")

Think of a car engine.

• Traditional Testing: Checks if the car can drive 60 mph. If yes, the mechanic says, "You're fine."
• This New Method: Listens to the engine. Even if the car hits 60 mph, the mechanic hears it sputtering, revving high, then stalling. The mechanic says, "The engine is unstable. It's going to break down soon."

The Takeaway:
This study suggests that cognitive inconsistency is a red flag. It's a sign that the brain's "top-down control" (the part that keeps you steady) is starting to fray, even before the person feels like they are failing at tasks.

By measuring this "wobble," doctors might be able to:

• Detect Parkinson's much earlier.
• Predict who will get worse faster.
• Test new drugs on people who are truly at risk, before the damage is done.

The Caveat (The "But...")

The authors are careful to say this is a preprint (a draft) and needs more testing. It's like finding a new type of weather radar. It looks promising, but we need to make sure it works in different cities (different groups of people) and doesn't get confused by other things like age or education.

In short: If your brain is a drum, this study found that the people with early Parkinson's aren't just playing the drum softer; they are playing it with a shaky, inconsistent rhythm. Catching that shaky rhythm early could change how we treat the disease.

Technical Summary

1. Problem Statement

Neuronal $\alpha$-synuclein disease (NSD), which includes Parkinson's disease (PD), involves pathological processes that begin years before overt clinical symptoms appear. The recently proposed NSD-ISS (Neuronal Synuclein Disease Integrated Staging System) defines Stage 2 as a prodromal phase characterized by the presence of pathological $\alpha$-synuclein and subtle clinical signs (e.g., hyposmia, REM sleep behavior disorder) but without functional impairment.

Current challenges in this field include:

• Early Detection: Differentiating Stage 2 participants from healthy controls (HCs) when mean cognitive performance remains within normal limits.
• Progression Prediction: Identifying biomarkers that can predict the transition from prodromal Stage 2 to more advanced stages (Stage 3+, involving functional impairment) before traditional mean-level cognitive decline is detectable.
• Biomarker Gap: There is a need for non-invasive, scalable cognitive metrics that reflect diffuse neuropathological changes (specifically in executive networks) earlier than standard neuropsychological averages.

2. Methodology

Study Design & Data Source

• Data Source: The Parkinson's Progression Markers Initiative (PPMI) database, accessed January 21, 2025.
• Participants: A total of 934 participants were analyzed:
  • NSD Stage 2 Cohort ($N=832$): Classified based on positive CSF $\alpha$-synuclein seed amplification assay (SAA) and/or DaTscan (dopaminergic dysfunction). Subdivided into Stage 2A (no dopaminergic dysfunction) and 2B (with dysfunction).
  • Healthy Controls ($N=102$): Required negative SAA, negative DaTscan, no family history of PD, and MoCA $\ge$ 27.
• Exclusion: Participants with Stage 3+ (functional impairment) or those missing the full cognitive battery were excluded.

Measures & Metrics

• Cognitive Battery: 11 neuropsychological tests covering memory, language, executive function, and processing speed (e.g., HVLT-R, WAIS-IV LNS, SDMT, TMT).
• Intraindividual Variability (IIV-D): Calculated using the Coefficient of Variation (CoV).
  • Total CoV: Standard deviation of a participant's 11 normed T-scores divided by their mean T-score.
  • Attention/Executive CoV: Calculated similarly but restricted to 5 specific executive/attention tests.
  • Note: Scores were converted to T-scores (Mean=50, SD=10) using internal PPMI norms adjusted for age, sex, and education.
• Clinical Correlates: MDS-UPDRS (motor), SCOPA-AUT (autonomic), GDS-15 (depression), STAI (anxiety), MoCA, and sleep/olfactory scales.
• Outcome Variable: Conversion from NSD Stage 2 to Stage 3+ within one year.

Statistical Analysis

• Group Comparisons: Independent t-tests (or Welch's t-tests) and Chi-square tests to compare Stage 2 vs. HCs, and Converters vs. Non-converters.
• Correlation: Pearson correlations between IIV-D metrics and clinical variables.
• Predictive Modeling: Binomial logistic regression to assess if baseline IIV-D predicts 1-year conversion, controlling for baseline motor severity (UPDRS Part III).
• Correction: Benjamini-Hochberg (BH) procedure applied to control the False Discovery Rate (FDR) at 0.05.

3. Key Contributions

1. Validation of IIV-D in Prodromal NSD: Demonstrates that cognitive dispersion (variability) is elevated in Stage 2 NSD patients compared to healthy controls, even when mean cognitive scores are normal.
2. Differentiation of Stages: Shows that IIV-D metrics can distinguish between stable prodromal patients and those who will rapidly progress to functional impairment within one year.
3. Independence from Motor Symptoms: Establishes that IIV-D is an independent predictor of disease progression, distinct from motor severity.
4. Clinical Utility: Proposes CoV as a scalable, non-invasive biomarker for early risk stratification in synucleinopathies.

4. Results

Demographics & Baseline Characteristics

• Stage 2 participants were significantly older than HCs but matched on education and gender.
• Stage 2 participants showed significantly worse motor scores, autonomic dysfunction, depression, anxiety, and olfactory function compared to HCs.
• Crucially: Mean cognitive performance (Composite T-score) for Stage 2 was 47.56 (within the normal range of 40–60), confirming that mean-level decline was not yet present.

Group Differences (Stage 2 vs. Healthy Controls)

• Total CoV: Stage 2 participants had significantly higher Total CoV ($p < .001$, $d = -0.31$) compared to HCs.
• Attention/Executive CoV: No significant difference was found between groups for this specific domain metric.
• Correlations: Total CoV was positively correlated with age, motor severity (UPDRS III), depression, and anxiety, and negatively correlated with education and olfaction.

Progression Analysis (Converters vs. Non-Converters)
Among Stage 2 participants, 100 individuals (12%) converted to Stage 3+ within one year.

• Baseline Differences: Converters had significantly higher Total CoV ($p = .008$) and Attention/Executive CoV ($p = .020$) at baseline compared to non-converters.
• Motor Symptoms: Converters had significantly higher baseline motor severity (UPDRS III) and Hoehn & Yahr staging.
• Cognitive Performance: Converters had lower mean composite scores, though still within normal limits.

Predictive Modeling

• Total CoV Model: Total CoV was a significant independent predictor of 1-year conversion ($OR = 1.44$, $95\% CI: 1.10–1.88$, $p = .008$) after controlling for motor severity.
  • Interpretation: For every unit increase in CoV, the odds of progressing to Stage 3+ increased by 44%.
• Attention/Executive CoV Model: Also a significant predictor ($OR = 1.38$, $p = .023$), though the effect size was slightly smaller.
• Model Fit: Both models explained approximately 15% of the variance in conversion status.

5. Significance & Implications

• Early Detection Mechanism: The study suggests that cognitive inconsistency (variability) emerges earlier than cognitive decline (mean score drop). Patients may appear "normal" on average but exhibit "noisy" performance across domains, signaling early network disruption (likely in frontal/executive systems).
• Clinical Application: Incorporating IIV-D metrics (specifically CoV) into routine cognitive screening could enhance the detection of prodromal NSD and identify high-risk individuals for clinical trials or early intervention.
• Pathophysiological Insight: Elevated IIV-D likely reflects diffuse neuropathological changes and diminished top-down executive control, serving as a proxy for broader disease progression beyond just motor or specific cognitive deficits.
• Limitations & Future Directions:
  • The study relies on a single cohort (PPMI); replication in independent populations is needed.
  • The follow-up period was short (1 year); long-term trajectories need assessment.
  • Future work should investigate differences between Stage 2A and 2B and explore neurobiological correlates (e.g., imaging, fluid biomarkers) of IIV-D.

Conclusion: Intraindividual cognitive variability, quantified by the Coefficient of Variation, is a sensitive, independent biomarker for differentiating early NSD from healthy aging and predicting short-term disease progression to functional impairment, offering a promising tool for early detection strategies.