Abstract
BACKGROUND AND PURPOSE: Inter-individual variability in levodopa responsiveness complicates personalized treatment of Parkinson's disease (PD). Previous studies have suggested that magnetic resonance imaging (MRI)-based indices of the perivascular space (PVS) and choroid plexus (CP) volume may be related to levodopa responsiveness. This study integrated multiple MRI indices, including CP volume, free water (FW) fraction, PVS volume fraction (PVSVF), and diffusion tensor imaging along the PVS (DTI-ALPS), to investigate their associations with levodopa responsiveness. METHODS: This retrospective study included 100 participants with PD (median age, 63.5 years; 53% females) who underwent 3T MRI between March 2023 and December 2024 and were grouped into good (n = 54) and poor (n = 46) responders based on the results of an acute levodopa challenge test. CP volume, FW fraction, PVSVF, and DTI-ALPS index were calculated. The Mann-Whitney U test and binary logistic regression were used for analysis. RESULTS: Participants in the poor responder group had a higher CP volume (p = 0.048) and FW fraction (p < 0.01) than those in the good responder group. Higher CP volume (odds ratio [OR], 0.986; p = 0.038) and FW fraction (odds ratio [OR], 0.883; p < 0.01) were significantly associated with poor levodopa responsiveness, while higher PVS volume fraction, higher PVSVF-BG, higher PVSVF-WM, and lower DTI-ALPS were not. CONCLUSION: Higher CP volume and FW fraction were independently associated with poor levodopa responsiveness. CP volume and FW fraction measurements may be valuable imaging biomarkers for predicting levodopa responsiveness.