Neural Mechanisms of Monoaminergic Engagement in Late-life Depression Treatment Response (NEMO)

In this competing renewal (Year 11) of the investigators' R01 which has used functional
magnetic resonance imaging (fMRI) to study late-life depression (LLD) pharmacotherapy
(R01MH076079), the primary aim of this study is to characterize functional connectivity
changes associated with initial medication exposure (12-hour challenge). Preliminary data
suggests that these initial fMRI changes reflect monoaminergic engagement, regardless of
monoaminergic class, and predict later treatment response. This study will test a neural
systems level model that response in LLD is mediated by acute pharmacologically-induced
changes in cognitive and affective large scale network.

Depression in older adults is frequently disabling and is often resistant to first-line
treatments, requiring more prolonged treatment trials than in younger adults, mainly due to
its heterogeneous pathophysiology (e.g. vascular and degenerative brain changes). Currently,
there is little neurobiological data to guide changing or augmenting antidepressant
medications. Thus, there has been a heightened focus on tailoring treatment to optimize
outcome as described in the 2015 National Institute of Mental Health (NIMH) draft strategic
plan (strategy 3.2). While antidepressant clinical response may take up to 8 weeks, recent
studies suggest that physiologic changes, as measured with pharmacologic fMRI (phMRI) are
seen within 12 hours of starting a new monoaminergic antidepressant (1).

For this proposal, investigators focus on three major Cognitive and Affective Networks (CAN):
the Default Mode Network (DMN), the Salience Network (SN) and the Executive Control Network
(ECN). The proposed model suggests that monoaminergic engagement leads to core CAN changes,
changes that subsequently are related to overall clinical response as well as response in
specific symptom clusters such as negative bias, somatizations/anxiety and cognitive control.
The same networks that are functionally connected while individuals are at rest, are also
selectively engaged during tasks. Investigators' prior work shows that pharmacotherapy -
regardless of type of antidepressant used - engages these specific networks at rest and
during standard cognitive and affective tasks. Given the role of cerebrovascular disease in
LLD treatment response, the moderating role of vascular burden on the proposed association
between CAN engagement and treatment response will also be explored.

The University of Pittsburgh will recruit 100 older adults with LLD that will be randomized
to receive treatment with either a very specific serotonin reuptake inhibitor (escitalopram)
or a norepinephrine reuptake inhibitor (levomilnacipran). A pair of fMRI scans one day apart
will be used to measure functional connectivity (FC) associated with medication titration.
Investigators will use a very early (12 hours after initiation of treatment) biomarker of
treatment response, which, when validated, would decrease substantially the waiting time
between medication changes. Additionally, this study will further increase knowledge of the
acute neural system changes associated with monoaminergic antidepressants; this knowledge of
mechanism is essential for both guiding LLD treatment research, and serving as an engagement
target in LLD treatment research.

Note: The original study design involved randomization to escitalopram or placebo (instead of
escitalopram and levomilnacipran). Therefore a subset of participants will complete the study
according to this design.

Inclusion Criteria:

- Age greater than or equal to 60 years old

- Major Depressive Disorder and current Major Depressive Episode

- Montgomery-Asberg Depression Rating Scale (MADRS) greater than or equal to 15

- Modified Mini-Mental State (3MS) score greater than or equal to 84

Exclusion Criteria:

- History of Mania or Psychosis

- Current suicidal ideation that cannot be safely managed within the confines of a
clinical trial

- Alcohol or Substance Abuse (current or past 3 months) endorsed via phone screening
interview or diagnosed by Structured Clinical Interview for the Diagnostic and
Statistical Manual for Mental Disorders (SCID)

- Dementia of any etiology endorsed via phone screening interview or diagnosed by SCID

- Medical conditions with known significant effects on mood (e.g., stroke, current
hypothyroid state) as well as unstable medical illness, including delirium,
uncontrolled diabetes mellitus, hypertension, hyperlipidemia, or cardiovascular risk
factors that are not under medical management

- Unwilling or clinically determined to be unable to taper from high doses of
benzodiazepines (equivalent to > 2 mg lorazepam/day) or other
anti-depressant/anti-anxiety medications at time of screening

- Inability to complete required assessments including brain MRI and blood draw

- Hearing/vision impairment precluding neuropsychological testing

- Difficulty conversing in English

- Clinical contraindication to use of escitalopram or levomilnacipran or history of
treatment resistance to escitalopram or levomilnacipran

- Unable or unwilling to provide a secondary/emergency contact person

- Confirmed history of stroke, current epilepsy, or current post-concussive symptoms

- Impaired renal function defined as a glomerular filtration rate (GFR) of 60 or lower