One patient, one destiny: A cluster analysis of the Parkinson's progression Markers Initiative (PPMI) cohort.
Longitudinal cluster analysis of 209 PPMI PD patients identified slow, intermediate, and rapid motor progression trajectories, found higher baseline BMI linked to faster motor decline, and showed that detailed baseline clinical and genetic/SAA profiles poorly predicted individual outcomes.
What the AI sees
Longitudinal cluster analysis of 209 PPMI PD patients identified slow, intermediate, and rapid motor progression trajectories, found higher baseline BMI linked to faster motor decline, and showed that detailed baseline clinical and genetic/SAA profiles poorly predicted individual outcomes.
Research significance
The study underscores profound individual variability that complicates clinical trial design and patient stratification and flags BMI as a potential modifiable factor to investigate, but it offers limited direct mechanistic or therapeutic leads.
Source abstract
BACKGROUND: Parkinson's disease (PD) is characterized by heterogeneity. While previous efforts have aimed to categorize patients by clinical subtypes, these classifications have not reliably predicted outcomes, underscoring the complexity and variability of disease progression. METHODS: Using data from the Parkinson's Progression Markers Initiative (PPMI), we conducted a longitudinal analysis of 209 individuals with PD over five years. We used cluster analysis to identify trajectory clusters in the progression of motor symptoms. Participants were categorized based on baseline variables, including age, sex, race, BMI, genetic mutation status, medical history, presence of jaw and limb tremors, and alpha-synuclein seeding status. Motor and non-motor outcomes, such as gait abnormalities, dyskinesia, motor fluctuations, dystonia, cognitive impairment, hallucinations, and others, were evaluated, along with the progression of the levodopa equivalent daily dose (LEDD). Sankey plots illustrated patient trajectories from baseline to outcome groups over time. RESULTS: Trajectory cluster analysis of MDS-UPDRS Part III scores identified three progression patterns: slow, intermediate, and rapid, with higher baseline BMI associated with faster motor progression (P < 0.01). Grouping based on baseline features identified 136 distinct patient groups, with substantial dispersion across clinical outcomes. Patients with identical baseline profiles frequently diverged into separate outcome groups. Genetic/SAA stratification also showed marked variability in progression with no significant association across trajectory clusters (P > 0.05). CONCLUSION: These findings highlight the existence of highly individualized clinical trajectories, even among patients with similar baseline characteristics, within the specific set of baseline variables examined in this study. This emphasizes the need for a personalized, patient-centered approach in both the management and study of Parkinson's disease.