Objective:

This work used established methods to generate a composite scale from MDS-UPDRS-Part II and PDQ-39 items optimized for sensitivity to detect functional decline over 1-year.

Background:

Traditional measures used in clinical trials of disease-modifying treatments (DMTs) in early Parkinson’s disease (PD) may fail to detect treatment effects on patients’ daily functioning over feasible study timeframes for otherwise effective treatments.

Design/Methods:

Data from 140 subjects with confirmed early PD who were naïve to dopaminergic therapies were selected from the placebo groups of five DMT trials within the Critical Path for Parkinson’s dataset. Patients were censored at PD treatment initiation. Partial least squares regression was used to select and weight items from MDS-UPDRS-Part II and PDQ-39 that maximize the observable signal of disease progression in early unmedicated PD (PARCOMS-Function). PARCOMS-Function’s ability to detect change was measured using a 1-year mean-to-standard-deviation ratio (MSDR), a ratio of signal-to-noise with higher values indicating better sensitivity.

Results:

This method selected 15 of the 52 possible items to form PARCOMS-Function – seven from MDS-UPDRS-Part II and eight from PDQ-39. The items with the highest contribution (indicating responsiveness to change in this population) were hygiene (17.2%), speech (12.6%), and handwriting (11.2%) from Part II and walking half a mile (8.7%), muscle cramps (8.3%), and buttons and shoelaces (6.3%) from PDQ-39. The 1-year MSDR of PARCOMS-Function was 0.6534, reflecting increases in the ability to detect change of 12.3% compared to Part II alone, and 339.1% compared to PDQ-39 alone.

Conclusions:

PARCOMS-Function is a re-weighted optimized version of existing measures, with greater sensitivity to detect functional decline in early unmedicated patients with PD. Use of PARCOMS-Function could increase trial efficiency and power to detect meaningful delay in disease progression with DMTs.