Late-Life Depression (LLD), or depression in older adults, often presents with motivational deficits, deficits in performance in cognitive domains including processing speed and executive dysfunction, and mobility impairments. This triad of findings implicate dopaminergic dysfunction as a core pathophysiologic feature in depression, and may contribute to cognitive decline and motor disability. Normal aging results in brain-wide dopamine declines, decreased D1/D2 receptor density, and loss of dopamine transporters. Although brain changes associated with depression and aging converge on dopamine circuits, the specific disturbances in LLD and how responsive the system is to modulation remain unclear. In this study, investigators are testing integrative model that aging, in concert with pro-inflammatory shifts, decreases dopamine signaling. These signally changes affects behaviors supported by these circuits, in the context of age-associated cortical atrophy and ischemic microvascular changes, resulting in variable LLD phenotypes. Investigators propose a primary pathway where dopaminergic dysfunction in depressed elders contributes to slowed processing speed and mobility impairments that increase the effort cost associated with voluntary behavior. The central hypothesis of this study is that late-life depression is characterized by dysfunction in the dopamine system and, by enhancing dopamine functioning in the brain. By improving cognitive and motor slowing, administration of carbidopa/levodopa (L-DOPA) will improve depressive symptoms.

Late Life Depression (LLD) is a prevalent, disabling, and at times lethal condition for which currently available treatments are often ineffective. No prior study has comprehensively examined dopamine-dependent behaviors (i.e., reward processing, cognition, motor function) in LLD, and none has integrated positron emission tomography (PET), multimodal magnetic resonance imaging (MRI), neuropsychological evaluation, and mobility assessments. Should cognitive and motor slowing result in altered effort-based decision making as researchers hypothesize, treatment development may shift from addressing mood and hedonic responses toward facilitating cognition and movement, reducing the effort cost of voluntary behavior, and promoting behavioral activation.

This study, across collaborative sites, will enroll 100 evaluable elderly depressed outpatients who enter baseline study procedures and who exhibit evidence of dopaminergic dysfunction, characterized as either slowed processing speed or slowed gait speed. To disentangle depression effects from age-related changes, 70 never-depressed elders also will complete baseline evaluation. To achieve this goal, at Vanderbilt University Medical Center (VUMC) the investigators will enroll 80 depressed elders and 50 never-depressed elders who complete baseline study procedures. The University of Pittsburgh Medical Center will enroll an additional 20 depressed elders and 20 never-depressed elders who complete baseline study procedures.

Assessments include PET imaging of receptor density, neuromelanin-sensitive MRI (NM-MRI) measurement of nigrostriatal status, task-based MRI focused on effort-based decision making and reward processing, and comprehensive psychiatric, neurocognitive, and physical performance evaluation. Depressed participants then will be randomized to levodopa (L-DOPA) or placebo for 3 weeks, followed by repeat multimodal MRI and cognitive/behavioral assessments. In a cross-over phase, participants will receive the opposite intervention for an additional 3 weeks followed by clinical and cognitive assessments only. This mechanistic probe allows the investigators to examine the contributions and interrelationships of dopamine-dependent processes in LLD and evaluate the responsivity of dopamine systems in LLD to pharmacological stimulation.

AIM 1: To characterize dopaminergic dysfunction in LLD at molecular, circuit, and behavioral levels.

Hyp 1: Compared to age- and gender-matched controls on baseline functional MRI (fMRI), LLD participants will be less willing to expend effort for rewards and exhibit lower prefrontal cortex and striatal activation on the Effort Expenditure for Rewards Task (EEfRT). Hyp 2: Across all participants, lower striatal [18F]-FDOPA relative influx rate, lower midbrain & striatal [18F]-fallypride binding, and lower NM-MRI signal in the substantia nigra, pars compacta will predict lower performance across RDoC domains: Positive Valence (impaired willingness to expend effort, decreased neural activations on the EEfRT), Cognitive (slowed processing speed and executive dysfunction), and Sensorimotor (slowed gait speed). Hyp 3: Across all participants, slowed processing and gait speed likewise will predict lower willingness to expend effort on the EEfRT.

AIM 2: To examine responsivity of dopamine circuits in LLD to stimulation with L-DOPA.

Hyp 1: Compared to placebo, L-DOPA will result in greater normalization of neural activations and improved behavioral performance in Positive Valence, Cognitive, and Sensorimotor domains. Hyp 2: Baseline PET and NM-MRI measures will moderate L-DOPA effects. The greatest improvements will be observed in those with the lowest baseline [18F]-FDOPA relative influx rate, [18F]-fallypride binding, and NM-MRI signal.

Exploratory Aims: 1) To investigate associations of baseline proinflammatory markers with dopaminergic function across molecular, circuit, cognitive and behavioral levels of analysis. 2) To evaluate the durability of L-DOPA effects on RDoC domains in the crossover phase.

Inclusion Criteria:

1. Age ≥ 60 years
2. Diagnostic and Statistical Manual-5 (DSM5) diagnosis of Major Depressive Disorder, Persistent Depressive Disorder, or Depression Not Otherwise Specified (NOS)
3. MADRS score ≥ 15
4. Decreased processing speed (0.5 SD below age-adjusted norms on the WAIS-IV Coding task or Trail Making Test, Part A) or decreased motor speed (gait speed/average walking speed on 15' course ≤ 1m/s, or 0.5 SD below age-, gender- and education-adjusted norms on the grooved pegboard test)
5. Capable of providing informed consent and adhering to study procedures

Exclusion Criteria:

1. Diagnosis of substance abuse or dependence (excluding Tobacco Use Disorder) in the past 12 months
2. Other psychiatric disorders including a history of psychosis, psychotic disorder, mania, or bipolar disorder. Other comorbid psychiatric disorders are allowable if the depressive disorder diagnosis is considered to be the primary problem
3. Primary neurological disorder, including dementia, stroke, Parkinson's disease, epilepsy, etc
4. SBT > 10
5. MADRS suicide item > 4 or other indication of acute suicidality
6. History of inpatient psychiatric hospitalization in the last year;
7. History of suicidal ideation in the last 6 months, operationalized as a 'yes' response to item 4 or 5 in the "Suicidal Ideation" section of the Columbia-Suicide Severity Rating Scale (CSSRS)
8. Any suicidal behavior in the last year (operationalized as a 'yes' response to any item in the "Suicidal Behavior" section of the CSSRS, including actual interrupted, aborted, or preparatory acts)
9. Current or recent (within the past 2 weeks) treatment with antipsychotics or mood stabilizers, or use of antidepressants where washout is not advisable
10. History of hypersensitivity, allergy, or intolerance to Carbidopa/levodopa
11. Any physical or intellectual disability adversely affecting ability to complete assessments
12. Unstable medical illness
13. Mobility limiting osteoarthritis of any lower extremity joints, symptomatic lumbar spine disease, or history of joint replacement / spine surgery that limits mobility
14. Diagnosis of HIV
15. History of significant radioactivity exposure (nuclear medicine studies or occupational exposure).

This study will include clinical, cognitive, and neuroimaging data from older depressed subjects. The final dataset will include clinical information about subject psychiatric diagnoses, psychiatric and medical history, cognitive data, and response to L-dopa . The investigators will share data via the National Database for Clinical Trials related to Mental Illness (NDCT). NDCT provides a secure platform for data-sharing allowing for communication of research data, tools, and supporting documents. As required by NDCT, the investigators will obtain a Global Unique Identifier (GUID) for each participant. The investigators will additionally follow NDCT requirements to certify and review data, as well as timeline requirements for data submission and data sharing. Sharing of neuroimaging data will also be facilitated by an XNAT system (xnat.org). XNAT is an open-source informatics software platform that assists in the management and archiving of imaging data.

Data will be shared according to policies from the NDCT and NDA (NIMH Data Archive). Descriptive data, outcome measures and analyzed data will be shared will be shared within 4 months of when a publication is accepted. Study data will be shared through the NDCT indefinitely.

The NIH will provide access to scientific investigators for research purposes. Qualified researchers who have completed a Data Use Certification and received approval from the NDA Data Access Committee (DAC) may be approved to access broadly shared data. A separate request process exists for access to data in federated sources. Additionally, the DAC and support staff at NIH have access to NDA shared data.

Double Blinded