ORIGINAL RESEARCH article
Front. Neurosci.
Sec. Neurodegeneration
Factors Associated with Longitudinal MDS-UPDRS III Score Trajectories in Early-Stage Parkinson's Disease
Provisionally accepted
- Chengdu Second People's Hospital, Chengdu, China
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
Background: Parkinson's disease (PD) exhibits significant clinical heterogeneity, particularly in motor symptom progression. This study aims to identify distinct trajectories of motor progression in PD and explore associated predictive factors. Methods: Data were obtained from the Parkinson's Progression Markers Initiative (PPMI) database on [2025-3-25]. Motor symptom severity was measured using the MDS-UPDRS III scores. Latent class trajectory analysis was used to identify distinct progression patterns. Multinomial logistic regression and machine learning models were used to evaluate predictors. Results: Three distinct motor progression trajectories were identified: slow progression (38 %), moderate progression (55.9 %), and rapid progression (6.1 %). Compared to the slow progression group, a higher baseline MDS-UPDRS III score was strongly associated with both moderate (OR = 1.27, 95% CI: 1.23–1.31, p < 0.001) and rapid progression (OR = 1.49, 95% CI: 1.43–1.57, p < 0.001). Lower serum albumin levels also significantly increased the likelihood of moderate (OR = 0.95, 95% CI: 0.91–0.99, p = 0.014) and rapid progression (OR = 0.89, 95% CI: 0.81–0.98, p = 0.016). Additionally, higher baseline BMI (per 5 kg/m2 increase) was associated with greater odds of moderate (OR = 1.19, 95% CI: 1.01–1.41, p = 0.042). Finally, each 1-unit lower mean striatum specific binding ratio (SBR) reduced the odds of moderate progression by 68 % compared with the slow-progression group (OR = 0.68, 95 % CI: 0.46–0.99, P = 0.044). Machine learning analysis confirmed the predictive importance of these factors, with the Random Forest model achieving an AUC of 0.950. Conclusion: Baseline motor severity, dopaminergic imaging, nutritional status, and body weight are key predictors of motor progression in PD. These findings highlight the potential for early risk stratification and personalized management strategies.
Keywords: machine learning, Movement Disorder Society - Unified Parkinson's Disease RatingScale Part III score, Parkinson's disease, Parkinson's ProgressionMarkers Initiative, Trajectories analysis
Received: 02 Dec 2025; Accepted: 09 Feb 2026.
Copyright: © 2026 Zhou, Liu, Zeng and Xia. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Wen Zhou
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.