Clinical misdiagnosis could be common across Parkinsonian disorders in a large autopsy-confirmed study, with researchers reporting movement disorder diagnostic error rates of approximately 10% to 20%, despite established clinical criteria. The findings showed that genetic variation, co-pathology, and ancestry could be associated with differences in underlying pathology that were not always reflected in clinical diagnoses.
In the study, the researchers analyzed clinical, neuropathologic, and genetic data from 3,353 patients and controls enrolled through 11 academic brain banks in the United Kingdom, United States, and Australia. The retrospective cross-sectional analysis included patients with Parkinson's disease, Parkinson's disease dementia, dementia with Lewy bodies, progressive supranuclear palsy, multiple system atrophy, and corticobasal syndrome as well as neurologically normal controls. Brain donations were collected between 1985 and 2024.
The researchers evaluated diagnostic accuracy by comparing clinical diagnoses with autopsy-confirmed pathology. Additional analyses examined Lewy body pathology, Alzheimer's disease co-pathology, survival, genetic variation, and ancestry-related differences in pathologic diagnoses.
Across the movement disorders included in the study, misdiagnosis rates ranged from approximately 10% to 20%. Clinical diagnoses of Parkinson's disease, Parkinson's disease dementia, and dementia with Lewy bodies corresponded with underlying Lewy body pathology in 92% of the cases. The researchers found that the donors with dementia-associated parkinsonism—Parkinson's disease dementia and dementia with Lewy bodies—were nearly twice as likely to have Lewy body pathology compared with those with Parkinson's disease without dementia.
Diagnostic performance varied across the disorders. Clinical diagnoses of multiple system atrophy and corticobasal syndrome showed high specificity but lower positive predictive value, reflecting the difficulty of predicting underlying pathology based on clinical presentation alone. Among patients with clinically diagnosed corticobasal syndrome, autopsy findings frequently revealed progressive supranuclear palsy or Alzheimer's disease pathology rather than corticobasal degeneration.
The researchers also reported substantial overlap between neurodegenerative pathologies. Among patients with Lewy body disease, 40% had coexisting Alzheimer's disease pathology. Greater Alzheimer's disease pathology was associated with more extensive Lewy body distribution. The burden of both Lewy body and Alzheimer's disease pathology increased across the clinical spectrum from Parkinson's disease to Parkinson's disease dementia to dementia with Lewy bodies. Neocortical Lewy body pathology was present in 53% of the patients with Parkinson's disease, 67% of those with Parkinson's disease dementia, and 81% of those with dementia with Lewy bodies.
Genetic analyses identified differences in pathologic burden among carriers of common Parkinson's disease–associated variants. Donors with GBA1 variants had more widespread Lewy body pathology compared with noncarriers and compared with those carrying LRRK2 variants. In contrast, those with pathogenic LRRK2 variants were less likely to exhibit advanced Lewy body pathology and experienced longer survival following disease onset compared with donors without known pathogenic variants.
The researchers also identified ancestry-related differences in pathology. Lewy body pathology was more common among patients of Ashkenazi Jewish ancestry, whereas progressive supranuclear palsy pathology was more frequently identified among those of South Asian ancestry. However, the findings may reflect recruitment patterns and brain bank representation in addition to biological differences, the researchers noted.
Several limitations should be considered when interpreting the findings. Because the study was based on brain bank cohorts, referral and sampling bias may have affected the results. The researchers also noted variability in the pathologic assessment across centers and time periods, incomplete documentation of some co-pathologies, and limited representation of non-European ancestry groups.
Overall, the findings suggested that clinical features alone do not fully capture the biological heterogeneity of Parkinsonian disorders.
“Findings of this cross-sectional brain bank study highlight the value of integrating genetic and pathologic data to improve diagnostic accuracy,” wrote lead study author Lesley Y. Wu, PhD, of the Department of Clinical and Movement Neurosciences at the University College London Queen Square Institute of Neurology, and colleagues. The researchers added that the results of the study support biologically informed diagnostic approaches that integrate neuropathology, genetics, and in vivo biomarkers.
Full disclosures of the study authors can be found in the study.
Source: JAMA Neurology