Send to

Choose Destination
J Neurol Sci. 2018 Apr 15;387:103-108. doi: 10.1016/j.jns.2018.01.025. Epub 2018 Jan 31.

Identification of a prospective early motor progression cluster of Parkinson's disease: Data from the PPMI study.

Author information

Department of Neurology, Athens Naval Hospital, Deinokratous, Athens 70, Greece. Electronic address:
Department of Neurology, Athens Naval Hospital, Deinokratous, Athens 70, Greece.
1st Department of Neurology, Eginition Hospital, National and Kapodistrian University of Athens, Greece.
Statens Serum Institut, 5 Artillerivej, Copenhagen DK-2300, Denmark.
Department of Physiology, Faculty of Medicine, University of Thessaly, Biopolis, Larissa 41110, Greece.
Division of Basic Neurosciences, Biomedical Research Foundation of the Academy of Athens, Athens, Greece; Second Department of Neurology, National and Kapodistrian University of Athens Medical School, Athens, Greece.



The aim of our study is to phenotype PD motor progression, and to detect whether serum, cerebrospinal fluid (CSF), neuroimaging biomarkers and neuropsychological measures characterize PD motor progression phenotypes.


We defined motor progression as a difference of at least one point in the Hoehn & Yahr (H&Y) scale between the baseline (Visit 0, V0), 12 months (Visit 04, V04) and 36 months (Visit 08, V08) milestones of the Progression Markers Initiative (PPMI) study. H&Y progression events were recorded at each milestone in order to be used as cluster analysis variables, in order to produce progression phenotypes. Subsequently, cross-cluster comparisons prior to and following (pairwise) propensity score matching were performed in order to assess phenotype - defining characteristics.


Four progression clusters where identified: SPPD: Secondarily Progressive PD, H&Y progression between V04 and V08; EPPD: Early Progressive PD. H&Y progression between V0 and V04; NPPD: Non Progressive PD, no H&Y progression; MIPD: Minimally Improving PD, i.e. Minimal H&Y improvement H&Y progression between V04 and V08;. Independent Samples Mann Whitney U tests determined CSF aSyn (p = 0.006, adj p-value = 0.036. I) and Semantic Animal fluency T-score (SFT, p = 0.003, adjusted p-value = 0.016.) as statistically significant cross-cluster characteristics. Following Propensity Score Matching, SFT, Hopkins Verbal Learning Test (Retention/Recall), Serum IGF1, CSF aSyn, DaT-SPECT binding ratios (SBRs) and the Benton Judgement of Line Orientation Test (BJLOT) were determined as statistically significant predictors of cluster differentiation (p < 0.05).


SFT, Serum IGF1, CSF aSyn and DaT-SPECT-derived, basal ganglia Striatal Binding Ratios warrant further investigation as possible motor progression biomarkers.


Biomarkers; Cluster analysis; Parkinson's disease; Phenotypes; Progression


Supplemental Content

Full text links

Icon for Elsevier Science
Loading ...
Support Center