Parkinson's disease subtypes: Approaches and clinical implications.

Abstract

Parkinson's disease (PD) is a complex neurodegenerative disorder with significant heterogeneity in disease presentation and progression. Subtype identification remains a top priority in the field of PD clinical research. Several PD subtypes have been identified. Hypothesis-driven subtypes refer to pre-defined subtypes based on specific criteria. Under hypothesis-driven subtypes, motor subtypes are the most common empirical subtype in both research and clinical settings. The concept of the non-motor symptoms (NMS) subtypes is relatively new and less well studied. Mild cognitive impairment (MCI) is one of the more prevalent NMS subtypes of PD. Data-driven subtyping is a hypothesis-free approach, that defines disease phenotypes by comprehensively evaluating multidimensional data. In this review, we summarize the main features for the different PD subtypes: from hypothesis-driven subtypes to data-driven subtypes. NMS and data-driven subtypes are still not yet well understood particularly with regard to biomarker and progression characterization. Future PD subtyping based on specific biological makers will enable us to better reflect the underlying pathophysiological underpinnings and enhance our search for specific therapeutic targets. The goal is to develop a simple algorithm to subtype PD patients at an early stage of PD that will enable good prognostication of their disease course, targeted therapies to be delivered, and proactive prevention of complications. Understanding PD subtypes and heterogeneity will also guide future clinical trial design and aid clinicians to better manage PD patients that will enable targeted disease surveillance and personalized treatment. The graphical abstract can be seen below.