Search for a command to run...
Parkinson's disease is a common and debilitating neurodegenerative disorder characterized by motor slowing (bradykinesia), which is thought to arise mainly owing to nigrostriatal dopaminergic cell loss. Paradoxically, longitudinal changes in striatal dopamine are poorly related to the progression of bradykinesia, indicating that other pathophysiological mechanisms play a role. In line with this, cross-sectional studies have shown that more benign motor phenotypes of Parkinson's disease are characterized by increased activity in the parieto-premotor cortex, indicative of cerebral compensation. However, the role of cerebral compensation in disease progression remains unclear. Here, we used a longitudinal design to test the hypothesis that the clinical progression of bradykinesia in Parkinson's disease is related to a decline in compensatory parieto-premotor function, over and above worsening nigrostriatal cell loss. We used a validated action selection task in combination with functional MRI to measure motor- and selection-related brain activity relative to the most-affected hand in a large sample of 351 patients with Parkinson's disease (≤5 years disease duration) and 60 healthy control subjects. In addition, we used diffusion-weighted MRI to obtain structural indices of substantia nigra and cerebral cortex integrity. These measurements were acquired at baseline and at 2-year follow-up, enabling us to compare longitudinal changes in brain metrics between patients and controls and to investigate their relationships with clinical metrics of bradykinesia progression. Consistent with our hypothesis, we observed that bradykinesia progression was inversely related to longitudinal changes in selection-related dorsal premotor cortex activity, suggesting that faster loss of cortical compensation contributes to faster symptom worsening. Importantly, this relationship remained after adjusting for longitudinal changes in the functional and structural integrity of the nigrostriatal system, indicating that bradykinesia progression is determined uniquely by loss of cortical compensation. In group comparisons of longitudinal change, patients with Parkinson's disease showed an overall reduction in putamen activity, which did not decrease further over time, in combination with an acceleration of structural decline in the substantia nigra and the premotor cortex. Despite showing expected patterns of Parkinson's disease pathology, neither of these metrics was correlated with bradykinesia progression. We conclude that the progression of bradykinesia in Parkinson's disease is determined by longitudinal changes in compensatory premotor cortex function. This presents opportunities to develop new progression-slowing interventions that focus on preserving and enhancing cortical compensation.