Objective: To evaluate how non-motor symptoms drive the classification of Parkinson’s disease (PD) patients beyond standard motor subtyping.
Background: While motor symptoms are the hallmark of PD, non-motor symptoms (NMS) often determine quality of life and progression. Understanding how NMS aggregate with motor phenotypes is essential for prognostic accuracy.
Method: Patients were evaluated using a comprehensive battery (UPDRS-III, NMSS, MMSE, FAB, SCOPA-Aut, RBD-SQ, and GDS/Beck) during the day-clinic at the department of Neurology of Sahloul University Hospital, Souse, Tunisia. Data were standardized (z-score), and dimensionality reduction was performed via Principal Component Analysis (PCA) to extract highly distinct feature components. These components were subsequently clustered using a K-means algorithm, with the optimal number of clusters determined via Silhouette Analysis and the Elbow Method.
Results: We included 58 PD patients (mean age=57.6, sex ratio=1.2). Clustering identified a Severe Non-Motorphenotype (Cluster 0) with significantly higher NMSS and SCOPA-Aut scores compared to other groups (p < 0.05). This group demonstrated a clear link between motor severity (mean off-state UPDRS-III = 52.4) and executive/cognitive dysfunction (Mean FAB score=9.9; Mean MMSE=22.7). In contrast, a Motor-Dominant/Mild group (Cluster 2) maintained high cognitive scores (mean MMSE= 28.1) despite motor symptoms. Cluster 1 represented a transition state where cognitive function was preserved, but autonomic and sleep symptoms (RBD-SQ) began to emerge alongside higher medication requirements.
Conclusion: Cognitive and autonomic symptoms are the strongest differentiators in patient clustering. The identification of a distinct cluster with severe NMS despite moderate disease duration suggests that NMS burden may represent a separate axis of disease progression independent of purely motor decline.
To cite this abstract in AMA style:
A. Rekik, R. Guizani, L. Bouafia, A. Mili, K. Jemai, H. Slimene, A. Hassine, S. Naija, S. Ben Amor. Non-Motor Symptom Burden as a Key Determinant of Disease Clustering in Parkinson’s Disease [abstract]. Mov Disord. 2026; 41 (suppl 1). https://www.mdsabstracts.org/abstract/non-motor-symptom-burden-as-a-key-determinant-of-disease-clustering-in-parkinsons-disease/. Accessed October 1, 2026.« Back to 2026 International Congress
MDS Abstracts - https://www.mdsabstracts.org/abstract/non-motor-symptom-burden-as-a-key-determinant-of-disease-clustering-in-parkinsons-disease/