Cognitive deficits are a core feature across disparate brain disorders, being highly prevalent and pervasive. Impairments in executive function are one of the most consistent findings in clinical and meta-analytical studies and were reported to be a principal mediator of psychosocial impairment and disability. Cognitive dysfunction is thought to be underlied by abnormalities in distributed brain circuits, at the cellular and molecular levels. Nonetheless, the neural mechanisms underlying the dysregulation in these circuits are poorly understood. Emerging evidence indicates that metabolic abnormalities are highly relevant for the domain of cognitive function and indicate that alterations in metabolic pathways may be relevant to neurocognitive decline across different populations. The incretin glucagon-like peptide-1 (GLP-1) is a hormone secreted by intestinal epithelial cells. GLP-1 receptors are widely expressed in the central nervous systems. Pre-clinical trials have demonstrated significant neuroprotective effects of GLP-1. Ongoing clinical trials measuring cognition and mood in populations with various psychiatric disorders lend further impetus to explore the effects of GLP-1R agonists on brain structure and cognitive function. We hypothesize that GLP-1 and the GLP-1R are relevant for molecular and cellular processes that are thought to underlie the formation and maintenance of brain circuits. A derivative of this hypothesis is that the administration of GLP-1 agonists may result in enhanced neuronal survival and consequential increase in gray matter volume. We therefore propose to explore the cellular and molecular abnormalities within and between neural circuits subserving cognition using the GLP-1R agonist liraglutide.

The overall goal of this study is to explore the relationship between a metabolic molecular target (i.e. the GLP1 system), the neural circuits of interest and the behavioral phenotype cognitive function.

We propose to explore the effects of GLP-1 agonism on brain structure and function. We hypothesize that the administration of GLP-1 agonists may result in increased connectivity in the executive control network. Multiple clinical trials with pharmacological agents, as well as cognitive therapy, have reported that measurements of brain structure and function are correlated with cognitive performance, indicating that they are a valid biological correlate of cognitive function. To assess this hypothesis we will recruit a clinical population, represented by individuals with a measurable impairment in executive function, wherein the target of our proposed intervention was shown to be altered.

We are currently not in a position to sufficiently homogenize subgroups of adults with mood disorder on the basis of any single or combinatorial biomarkers. It is also unlikely that alterations in GLP-1 receptor function, and the proposed model herein, is sufficiently explanatory to all sub-populations of adults with mood disorders. Instead, we propose that adults with mood disorders, who have co-existing metabolic disorders (e.g. type 2 diabetes mellitus), would be more likely to have a brain illness that is influenced by (i.e. cause, consequence or both) alterations in cellular bioenergetics. Moreover, convergent evidence suggest that GLP-1 receptor function may be, at least partially, dependent on glucose levels and/or insulin sensibility. It is a separate, yet testable, hypothesis that subpopulations enriched on the basis of having metabolic comorbidity (i.e. insulin resistance) may be more responsive to an intervention that targets a metabolic pathway.

We plan to test the effects of adjunctive liraglutide on executive function. We will select a subpopulation of patients, with a mood disorder and impairment in executive function, as defined by a below-average (i.e. 1 standard deviation below norm) performance in the Trail Making Test-B (TMTB). Furthermore, we plan to recruit two groups of patients, with and without insulin resistance, as defined by a homeostatic model assessment for insulin resistance (HOMA-IR) above 2.5, which will allow a comparison of the effects of liraglutide in a metabolically heterogeneous population.

Inclusion Criteria:

1. Informed consent before study-related activity
2. Individuals between the ages of 18 and 45 who meet DSM-5 criteria for bipolar disorders or depressive disorders
3. Below-average (i.e. 1 standard deviation below norm) performance in the TMTB.
4. Individuals must be receiving guideline concordant pharmacotherapy, in stable doses, without withdrawal or addition of medication in the last month.

Exclusion Criteria:

1. Diagnosis of possible or probable AD, MCI, or any other dementia
2. History of neurological disorder (ischemic attacks, carotid bruits, or lacunes upon MRI scan), or evidence of neurologic or other physical illness that could produce cognitive deterioration
3. Individuals in a severe mood episode, defined as a Hamilton Depression Rating scale 17- item (HAMD-17) total score of >23 or a Young Mania Rating Scale (YMRS) total score of >20.
4. Actively suicidal or evaluated as being a suicide risk (operationalized as a score of ≥3 on HAMD-17 suicide item and/or by clinical assessment).
5. Substance use disorder within 3 months before screening or a positive baseline toxicology screen.
6. Currently being treated for diabetes, with oral hypoglycemic agents and/or insulin, as these medications affect glucose and insulin levels, as well as the HOMA-IR calculation.
7. Presence of absolute or relative contraindication to liraglutide (e.g. hepatic impairment, renal impairment with CKD stage 3 and above, personal or familial history of medullary thyroid cancer or Multiple Endocrine Neoplasia syndrome type 2)
8. History of alcoholism; history of pancreatitis or pancreatic cancer
9. Presence of clinically unstable general medical illness.
10. Pregnancy or breastfeeding women.