Role of Inflammation on the Neurobiological Features of Depression & Potential Stratification

Project Summary:
Major depressive disorder (MDD) is a heritable and disabling psychiatric condition which is the leading cause of disability worldwide. MDD is attributable to the effects and interactions of genetic and environmental factors, although the pathophysiological mechanisms remain poorly understood. Previous imaging studies have reported alterations in brain regions and connecting tracts within affect- and reward-processing networks, but the direction of causality remains unclear (Whalley et al., 2013; Shen et al., 2017). Increasing evidence also suggests that MDD is not a unitary disorder but is likely to represent a grouping of conditions with diverse aetiologies with the resulting heterogeneity hampering progress.
There are increasingly compelling data implicating immune/inflammatory molecules in MDD pathophysiology (Leighton et al., 2017). There are also strong indications of a key role for early life adversity and related stress-HPA axis activation (Baumeister et al., 2016). These responses may have neurotoxic effects in the brain, ultimately resulting in the clinical symptoms and progression of the disorder. However, the effects of peripheral inflammation on brain structure and function in-vivo in the disorder remain poorly understood. Further, whether affected individuals with an inflammatory aetiology, or vulnerability, constitute a definable neurobiological sub-strata of MDD also remains unknown, but has strong implications for specific and mechanistically-tailored treatment.
This project will investigate the relationship between increased peripheral markers of inflammation (CRP, immunoglobulins, cortisol) and brain structure/function and will examine if these relationships can be used to sub-classify depression, potentially leading to targeted therapeutics. This study will use one of the largest single imaging studies of depression to date with neuroimaging and peripheral markers of immune-mediated inflammatory responses (STRADL n~1000, a subset of a larger cohort study Generation Scotland n=23,000 with in depth genotyping and phenotyping and access to birth cohort data). As part of this project we will test the relationship between peripheral blood markers collected at the time of imaging with structural brain connectivity (DTI- Diffusion Tensor Imaging), cortical thickness measures (structural MRI), subcortical brain volumes (hippocampus and striatum) and functional connectivity (resting-state fMRI). We will then apply machine learning techniques to attempt to sub-classify a neuroimmune subtype within the diagnosis of depression.
The hypotheses are that 1) increased inflammatory biomarkers will correlate with neuroimaging markers of decreased neuronal integrity and decreased functional connectivity between cortical and limbic areas, and abnormalities of brain structure, specifically reductions in hippocampal and striatal volume and 2) computerized classification techniques will identify a neurobiological and aetiologically homogeneous sub-group within the depressed cases.