Dopaminergic denervation and associated MRI microstructural changes in the nigrostriatal projection in early Parkinson's disease patients

Abstract

Loss of dopaminergic neurons in the substantia nigra pars compacta (SNc) and a profound reduction of striatal dopamine are two hallmarks of Parkinson's disease (PD). However, it's unclear whether degeneration starts at the neuronal soma or the striatal presynaptic terminals, and how microstructural degeneration is linked to dopaminergic loss is also uncertain. In this study, thirty de novo PD patients and twenty healthy subjects (HS) underwent 6-[¹⁸F]-fluoro-L-dopa (FDOPA) PET and MRI studies no later than 12 months from clinical diagnosis. FDOPA uptake rate (K_i), fractional volume of free-water (FW), and iron-sensitive R2* relaxometry were quantified within nigrostriatal regions. Inter-group differences (PD vs HS) were studied using non-parametric statistics and complemented with Cohen's d effect sizes and Bayesian statistics. Correlation analyses were performed exploring biomarker dependencies and their association with bradykinesia scores. PD patients exhibited a significant decline in nigrostriatal dopaminergic activity, being post-commissural putamen (-67%) and posterolateral SNc (-11.7%) the most affected subregions within striatum and SNc respectively. Microstructural alterations (FW) were restricted to the hemisphere corresponding to the most affected side and followed similar spatial gradients as FDOPA K_i (+20% in posterior putamen and +11% in posterolateral SNc). R2* revealed no relevant significant changes. FDOPA and FW were correlated within the posterolateral SNc, and clinical severity was associated with FDOPA K_i loss. The asymmetry between striatal and SNc changes for both dopaminergic depletion and microstructural degeneration biomarkers is consistent with a neurodegenerative process that begins in the striatal terminals before progressing toward the cell bodies in the SNc.

Fig. 1. Inter-group comparison of the imaging metrics.

Boxplot representation of the imaging metrics (FDOPA K_i, FW, and R2*) comparing HS and PD groups, in green and orange colors respectively. Boxplots are grouped by more affected side (MAS) and dominant side, and by less affected side (LAS) and non-dominant side. Inter-group statistical significances after Mann–Whitney U tests are displayed above boxplots as P > 0.10 (no symbol); 0.10 ≤ P < 0.05 (~); 0.05 ≤ P < 0.01 (*); P ≤ 0.01 (**). In boxplots, the center line denotes the median, lower and upper box lines respectively represent the first and third quartiles, whiskers the minimum and maximum values, and points the outliers.

Fig. 2. Nigrostriatal profile of change across imaging metrics.

Percent of change for FDOPA, FW, and R2* in all nigrostriatal divisions between healthy subjects (HS) and PD patients. More affected and less affected sides (MAS/LAS) were compared to the dominant and non-dominant sides respectively. Representation is visualized in an axial view with the striatum and the SNc represented as a surface mesh. Percentages are highlighted by color-coding. Positive and negative changes represent increments and reductions for the metrics in PD patients with respect to the HS cohort.

Fig. 3. Scatter plots between imaging metrics and clinical severity.

These scatter plots represent the correlation between FDOPA K_i and FW in the posterolateral SNc, and between FDOPA K_i in the posterior putamen and posterolateral SNc with the bradykinesia quantitative scores (Kinesia evaluation). The strength of these correlations (Spearman’s ρ) is reported within each scatter plot.

Fig. 4. ROI definition scheme.

a Anterior and posterior divisions of the striatum in both caudate and putamen structures in the T1w native space in one representative subject. The division was made with respect to the anterior commissure. b Anteromedial and posterolateral divisions of the substantia nigra pars compacta (SNc) with respect to the centroid of the red nucleus overlaid onto a T2w in standard ICBM-152 2009c space.