DESCRIPTION: Parkinson's disease (PD) is an incurable neurodegenerative disorder affecting approximately 1 million US adults and about 80,000 Veterans. PD causes significant morbidity due to motor and non-motor symptoms across its prolonged course with an annual economic burden of $14 billion in the US alone. Motor symptoms associated with loss of dopaminergic neurons in PD may be temporarily improved with dopamine replacing medicines, but disease-modifying therapies that delay or prevent neuronal loss are lacking and sorely needed. Exercise is promising as a disease-modifying therapy because it protects dopaminergic neurons in animal models of PD and has been associated with measures of neuroplasticity in PD patients. Unfortunately, more than half of dopaminergic neurons in the substantia nigra are lost before motor symptoms occur making it difficult to identify patients early enough to benefit from potentially disease-modifying therapies. Early "prodromal" PD can be identified using non-motor features including olfactory dysfunction and other biomarkers such as dopamine transporter (DaT) brain imaging abnormalities that are apparent years before motor symptoms. However, these strategies would be difficult and costly to implement on a population level without first identifying high-risk individuals for screening. Commonly prescribed dopamine blocking antipsychotic drugs cause debilitating PD-like motor dysfunction that is difficult to treat, and in some patients this findin may serve as a "stress test" for failing dopaminergic networks unmasking symptoms long before they would normally appear. Identifying prodromal PD among drug-induced parkinsonism patients offers a unique and unexplored opportunity for early intervention. In the proposed studies, we will employ a tiered screening strategy with inexpensive and non-invasive olfactory testing in drug-induced parkinsonism patients followed by DaT imaging in individuals with olfactory impairment to identify a cohort of patients with presumed prodromal PD. Subjects with presumed prodromal PD will then be randomized to a home-based exercise intervention ({5} times weekly aerobic walking confirmed by remote activity monitors) or no intervention. In this cohort, we will assess: 1) Short-term symptomatic effects of exercise on motor function in drug-induced parkinsonism using standard clinical measures (Unified Parkinson's Disease Rating Scale) and quantitative gait assessments after 8 weeks of intervention; 2) a potential disease-modifying effect after 52 weeks of exercise by comparing the rate of change in quantitative DaT imaging; and 3) the mechanisms and biochemical correlates of exercise-induced changes using a panel of serum biomarkers implicated in exercise and/or PD risk including brain-derived neurotrophic factor, uric acid, and apolipoproteinA1. Differences in the rate of change between groups will be assessed using independent samples t-tests and linear mixed-effects models adjusting...