Pathology and genetics in a global cohort of Parkinsonian Disorders

This multicenter retrospective study of 3,353 brain donors across 11 global brain banks reveals that misdiagnosis rates for parkinsonian disorders range from 10% to 20%, with significant associations found between specific genetic variants (GBA1, LRRK2), ancestry (South Asian and Ashkenazi Jewish), and distinct pathological burdens, underscoring the critical need for biologically informed diagnostic tools and genetically stratified approaches in future therapeutic trials.

The Gist

Imagine the human brain as a vast, complex city. Sometimes, this city gets clogged with "trash" (misfolded proteins) that causes traffic jams, leading to movement disorders like Parkinson's disease. For decades, doctors have tried to diagnose these traffic jams based on how the city looks from the outside (symptoms like shaking or stiffness). But this study, involving over 3,000 brain donors from around the world, suggests that looking from the outside isn't always enough. We need to look under the hood to see exactly what kind of trash is clogging the system.

Here is a simple breakdown of what the researchers found, using some everyday analogies:

1. The "Wrong Address" Problem (Misdiagnosis)

Think of the different brain diseases (Parkinson's, Lewy Body Dementia, Alzheimer's) as different types of construction projects. Sometimes, a doctor sees a building with a flat roof and assumes it's a warehouse (Parkinson's), but when they get inside, they find it's actually a library (Lewy Body Dementia) or a school (Alzheimer's).

• The Finding: In this study, about 10% to 20% of patients were given the wrong "address" while they were alive.
• The Clue: If a patient had memory loss (dementia) plus movement issues, the doctors were much better at guessing the right "address." It's like if you see a library card and books, you know it's a library, not just a warehouse. But if you only see the shaking hands, it's harder to tell which building it is.

2. The Genetic "Blueprints"

Every city has a blueprint (our DNA) that tells the construction crew how to build things. Sometimes, the blueprint has a typo. The study looked at two specific typos: GBA1 and LRRK2.

• The GBA1 Typo (The "Heavy Hauler"): People with this genetic typo tend to have a much heavier load of "trash" (Lewy bodies) in their brains. It's like their garbage trucks are overloaded, causing more widespread damage.
• The LRRK2 Typo (The "Slow Burn"): Surprisingly, people with this typo had less trash in their brains, but they lived longer. It's like their city is cleaner, but the construction is happening very slowly. They also had a different kind of "blueprint error" (Tau protein) that looked more like a different disease entirely (PSP).
• The Takeaway: Just because two people have the same shaking hands doesn't mean they have the same genetic blueprint or the same amount of trash in their brains.

3. The "Double Trouble" Effect

In many cases, the brain wasn't just dealing with one type of trash; it was dealing with two.

• The Finding: 40% of people with Lewy Body disease also had Alzheimer's disease "trash" (plaques and tangles) in their brains.
• The Analogy: Imagine a city dealing with a flood (Parkinson's) while simultaneously having a fire (Alzheimer's). The combination makes the situation much worse and leads to more severe symptoms, like memory loss.

4. The "Neighborhood" Factor (Ancestry)

The study looked at people from different backgrounds (ancestries) and found that the "trash" patterns varied depending on where the person's family came from.

• The Finding: People of Ashkenazi Jewish descent were more likely to have the "Lewy Body" trash, while people of South Asian descent were more likely to have the "Tau" trash (associated with Progressive Supranuclear Palsy).
• Why it matters: It's like different neighborhoods having different types of construction issues. If you only study one neighborhood, you miss the whole picture. We need to look at all neighborhoods to understand the city.

5. The "Normal" Controls

Even in people who were considered "neurologically normal" (no movement issues) when they were alive, the researchers found "trash" in 4% of their brains.

• The Analogy: It's like finding a few loose bricks in a house that looked perfectly fine from the outside. This suggests that the disease might start silently long before symptoms appear.

The Big Picture: Why This Matters

For a long time, we've been trying to fix the city based on a blurry photo of the street. This study says, "Let's get a high-resolution map of the underground pipes and the blueprints."

• Better Diagnosis: By combining genetics (the blueprint) with pathology (the actual trash), doctors can eventually diagnose these diseases more accurately while the patient is still alive.
• Better Treatments: If a drug is designed to clean up "Lewy Body trash," it won't work on a patient who actually has "Tau trash." This study helps researchers group patients correctly so they can test the right medicine on the right people.

In short: This massive study is like a global cleanup crew finally getting a detailed inventory of the mess in our brains. It tells us that the "one size fits all" approach to Parkinson's doesn't work. We need to know the specific genetic blueprint and the specific type of trash to fix the problem effectively.

Technical Summary

1. Problem Statement

Neurodegenerative movement disorders (e.g., Parkinson's disease [PD], Dementia with Lewy Bodies [DLB], Progressive Supranuclear Palsy [PSP], Multiple System Atrophy [MSA]) present significant diagnostic challenges due to:

• Clinical Overlap: Symptoms often overlap, especially in early stages, leading to high rates of clinical misdiagnosis.
• Lack of Biomarkers: There is a scarcity of reliable in vivo biomarkers to confirm underlying pathology before death.
• Genetic-Pathological Disconnect: While genetic variants (e.g., in GBA1, LRRK2) are known to influence disease risk, their specific impact on neuropathological burden and heterogeneity across diverse ancestries is not fully understood.
• Ancestry Bias: Previous research has predominantly focused on individuals of European ancestry, potentially overlooking ancestry-specific pathological differences.

2. Methodology

The study employed a multicentre, retrospective autopsy cohort design harmonizing data from 11 academic brain banks across the UK, US, and Australia.

• Cohort: 3,353 brain donors (1,281 female; mean age at death 76.8 years) with clinical diagnoses of PD, PD with dementia (PDD), DLB, PSP, Corticobasal Syndrome (CBS), MSA, or neurologically normal controls. Recruitment spanned 1985–2024.
• Data Integration:
  • Clinical: Antemortem primary diagnoses and disease duration.
  • Neuropathology: Post-mortem assessment of $\alpha$-synuclein (Lewy bodies), amyloid-$\beta$, and neurofibrillary tangles (NFTs) using standardized staging systems (Braak, CERAD, Thal, USSLB).
  • Genetics: All cases underwent genotyping (Illumina Neurobooster array) and/or genome sequencing. Variants were annotated using ClinVar and PanelAPP to identify pathogenic/likely pathogenic variants in genes associated with movement disorders (GBA1, LRRK2, SNCA, PRKN, TBK1, MAPT).
  • Ancestry: Genetically inferred ancestry was determined using Genotools.
• Statistical Analysis:
  • Diagnostic accuracy was measured via sensitivity, specificity, Positive Predictive Value (PPV), and Negative Predictive Value (NPV).
  • Associations between genetic status, ancestry, and pathological burden were assessed using proportional odds logistic regression (adjusted for sex, disease duration, and age at death) and Cox proportional hazards models for survival analysis.
  • Ancestry-pathology associations were tested using Pearson's Chi-square tests.

3. Key Contributions

• Scale and Diversity: This is one of the largest genotyped and genome-sequenced movement disorder cohorts integrated with neuropathology, including significant representation of non-European ancestries (e.g., South Asian, Ashkenazi Jewish).
• Genotype-Phenotype-Pathology Mapping: It provides a comprehensive map of how specific genetic variants (GBA1, LRRK2, APOE) modulate pathological burden (Lewy body distribution, Alzheimer's co-pathology) and survival.
• Ancestry-Specific Insights: It highlights distinct pathological distributions based on genetic ancestry, independent of known mutation status.
• Diagnostic Benchmarking: It establishes a robust "gold standard" for evaluating clinical diagnostic accuracy against post-mortem pathology in a global context.

4. Key Results

A. Diagnostic Accuracy

• Misdiagnosis Rates: Clinical misdiagnosis remained high (10–20%), consistent with historical data despite larger sample sizes.
• Dementia as a Predictor: Clinical diagnoses of dementia with parkinsonism (PDD/DLB) were nearly twice as likely to correspond to Lewy Body Disease (LBD) pathology compared to PD without dementia (OR = 1.96).
• Specific Syndromes:
  • CBS: Had the lowest PPV (23.7%), often misdiagnosed due to underlying PSP or Alzheimer's pathology.
  • MSA: High specificity (98.3%), but 16.8% of clinically diagnosed MSA cases had primary LBD pathology at autopsy.
  • Controls: 8.3% of clinically "normal" controls had background neurodegenerative pathology; 4.4% had Lewy body pathology (including neocortical involvement).

B. Co-pathology

• Prevalence: 84% of affected cases had co-pathology.
• Alzheimer's Disease (AD): AD co-pathology was present in 40% of LBD cases and was associated with more extensive Lewy body distribution (OR = 2.61).
• Impact: Co-pathology was associated with a later age of onset, suggesting an age-related decline in protein clearance mechanisms.

C. Genetic Associations

• GBA1: Carriers (both Gaucher-causing and PD-risk variants) exhibited significantly greater Lewy body burden (neocortical involvement) compared to non-carriers (OR = 1.94) and LRRK2 carriers (OR = 7.44).
• LRRK2:
  • Carriers (mostly p.G2019S) were enriched in Ashkenazi Jewish ancestry (21% vs. 1.7% in the rest of the cohort).
  • Pathologically, they had lower Lewy body burden and were less likely to have advanced LB stages compared to non-carriers.
  • Survival: LRRK2 carriers showed significantly longer survival (HR = 0.60) compared to non-carriers and GBA1 carriers.
  • Some LRRK2 carriers presented with PSP-like tau pathology or no pathology despite clinical PD.
• APOE $\epsilon$4: Each additional copy of the $\epsilon$4 allele doubled the odds of widespread LB pathology (OR = 2.25) in LBD cases.
• MAPT: H1/H1 haplotype was most frequent in PSP and least frequent in MSA.

D. Ancestry Differences

• Ashkenazi Jewish (AJ): Significantly more likely to have LBD pathology, independent of GBA1 or LRRK2 status.
• South Asian (SAS): Significantly more likely to have PSP pathology.
• Age: South Asian donors died at a significantly younger age than European or AJ donors.

5. Significance and Implications

• Clinical Practice: The findings underscore that clinical diagnosis alone is insufficient for precise stratification. The presence of dementia improves diagnostic accuracy for LBD, while specific genetic backgrounds (e.g., LRRK2, GBA1) predict distinct pathological trajectories.
• Therapeutic Trials: The study supports the need for pathology-informed and genetically stratified clinical trials. For instance, LRRK2 carriers may have different disease mechanisms (potentially oligomeric rather than fibrillar $\alpha$-synuclein) and survival profiles compared to sporadic PD, necessitating tailored therapeutic approaches.
• Biomarker Development: The high rate of misdiagnosis and the complexity of co-pathology highlight the urgent need for in vivo biomarkers (e.g., seed amplification assays for $\alpha$-synuclein) to differentiate these disorders during life.
• Global Representation: The study reveals that ancestry is a critical biological variable in movement disorders. Future research and brain bank recruitment must prioritize underrepresented populations to ensure diagnostic tools and therapies are effective across all demographics.

In conclusion, this study demonstrates that integrating genetics, ancestry, and detailed neuropathology is essential for unraveling the heterogeneity of Parkinsonian disorders, moving the field toward biologically defined disease classifications rather than purely clinical syndromes.