Does Abnormal Insulin Action in the Brain Underlie Cognitive and Metabolic Dysfunction in Schizophrenia

1. Schizophrenia and cognition: Cognitive impairment is a core aspect of schizophrenia (SCZ), contributing to disability and poor functional outcomes. Antipsychotics reduce positive symptoms but there are no currently approved treatments for cognitive impairment, creating a large unmet need. ii) Schizophrenia and metabolic dysfunction: Patients with SCZ also have exceedingly high rates of metabolic comorbidity. Almost half of patients are obese and the prevalence of type 2 diabetes is 3-9 fold higher than the general population. Patients with SCZ die on average 15-20 years earlier than the general population from cardiovascular disease. Thus, metabolic health represents another large unmet need. iii) Association between cognitive and metabolic dysfunction: These two domains of dysfunction interact in an additive manner to worsen outcomes. Metabolic syndrome and diabetes are both associated with worse cognition among SCZ patients. Recent knowledge elucidating the interactions between metabolic health, cognition, and functioning have encouraged a reconceptualization of SCZ as both a metabolic and cognitive disorder, prompting search for treatment strategies that address abnormalities in both these aspects. iv) Brain insulin as a unifying link: There has been recent recognition that insulin plays an important role in the brain. Brain insulin is implicated in several processes relevant to SCZ. Abnormal brain insulin action may help explain both cognitive and metabolic aberrations in patients with SCZ. Moreover, it is now clear that glucose uptake in the brain is partially dependent on insulin in brain regions relevant to SCZ, such as the hippocampus, hypothalamus, and striatum.
2. Evidence and promise in SCZ: There is preliminary evidence to suggest that brain insulin resistance is associated with worse cognition. A magnetic resonance spectroscopy (1H-MRS) study found higher brain glucose and lower glucose utilization in SCZ patients, suggesting brain insulin resistance, that were associated with memory impairment. Initial intervention studies using intranasal insulin have not been successful, likely because resistance to insulin in the brain prevents any benefits of intranasal insulin from accruing. However, this has not been conclusively demonstrated. This study seeks to answer this question directly. vi) Role of 18-fluorodeoxyglucose ([18F]-FDG)-positron emission tomography (PET): 1H-MRS is an indirect and imprecise measure of glucose in the brain (it combines intra- and extracellular glucose). This is also true for other MRI based measures, which have recently been employed to indirectly study insulin action in the brain. Currently, there are no PET ligands able to reliably quantify insulin or its receptors in the brain. However, using [18F]-FDG PET, it is possible to measure differences in glucose uptake, with and without an insulin challenge, into insulin sensitive regions of the brain (e.g. hippocampus and striatum). This can serve as a surrogate marker of brain insulin action. This principle has already been used successfully in rodents and healthy humans, and offers a more direct method of quantifying brain insulin action.

In this study, [18F]-FDG PET will be employed to examine whether abnormal brain insulin action is a feature of SCZ. The study examines young SCZ patients. Insulin (160 IU; shown to be safe and effective previously) will be delivered intranasally as it has been shown to be reliable method of delivering insulin to the brain.

Primary hypothesis: SCZ patients will have reduced [18F]-FDG uptake, in response to an intranasal insulin challenge, compared to healthy controls.