Establishment of a Biomarker-Directed Clinical Endpoint Model for Early-Stage Parkinson's Disease Patients

Abstract

Parkinson's Disease (PD) is a neurodegenerative disorder characterized by dopaminergic cell death in the substantia nigra. While the interplay between dopamine loss and symptoms is well-recognized, a respective quantitative link has yet to be established. The objective was to establish a biomarker-directed clinical endpoint model for early-stage PD patients. We developed a disease progression model using DATscan data in 196 healthy subjects and 419 Parkinson's patients to characterize the onset and progression of disease in early-stage PD patients. This disease progression model was then linked to MDS-UPDRS Parts I, II, and III data from the Parkinson's Progression Markers Initiative (PPMI) using a modified item response theory (IRT) analysis to characterize and predict the impact of dopamine loss on motor and non-motor symptoms. Disease onset occurs ~4-15 years pre-diagnosis. There is correlation (Spearman's rank correlation: 0.73-0.78, P < 0.001) between striatal binding ratio values (SBR) and MDS-UPDRS total scores in early-stage PD patients once interindividual differences in age at diagnosis and onset of symptoms are considered. Stratification by degree of damage improved the model's performance for putamen/motor symptoms but not for caudate/cognitive symptoms. The model captured changes in MDS-UPDRS Parts I, II, and III in early-stage, moderately progressing PD patients (60-65% of PPMI patients). In conclusion, we developed an SBR-directed IRT model that characterizes changes in MDS-UPDRS in > 60% of early-stage PPMI patients for ~15 years.