Comprehensively modeling heterogeneous symptom progression for Parkinson's disease subtyping

Parkinsons disease (PD) is a progressive neurodegenerative disorder marked by significant clinical and progression heterogeneity resulting from complex pathophysiological mechanisms. This study aimed at addressing heterogeneity of PD through the integrative analysis of a broad spectrum of data sources. We analyzed clinical progression data spanning over 5 years from individuals with de novo PD, using machine learning and deep learning, to characterize individuals phenotypic progression trajectories for subtyping. We discovered three pace subtypes of PD which exhibited distinct progression patterns and were stable over time: the Inching Pace subtype (PD-I) with mild baseline severity and mild progression speed; the Moderate Pace subtype (PD-M) with mild baseline severity but advancing at a moderate progression rate; and the Rapid Pace subtype (PD-R) with the most rapid symptom progression rate. We found that cerebrospinal fluid P-tau/-synuclein ratio and atrophy in certain brain regions measured by neuroimaging might be indicative markers of these subtypes. Furthermore, through genetic and transcriptomic data analyses enhanced by network medicine approaches, we detected molecular modules associated with each subtype. For instance, the PD-R-specific module suggested STAT3, FYN, BECN1, APOA1, NEDD4, and GATA2 as potential driver genes of PD-R. Pathway analysis suggested that neuroinflammation, oxidative stress, metabolism, AD, PI3K/AKT, and angiogenesis pathways may drive rapid PD progression (i.e., PD-R). Moreover, we identified candidate repurposable drugs via targeting these subtype-specific molecular modules and estimated their treatment effects using two large-scale real-world patient databases. The real-world evidence we gained revealed metformins potential in ameliorating PD progression. In conclusion, our findings illuminated distinct PD pace subtypes with differing progression patterns, uncovered potential biological underpinnings driving different subtypes, and predicted repurposable drug candidates. This work may help better understand clinical and pathophysiological complexity of PD progression and accelerate precision medicine.