University of Glasgow
Central consequences of peripheral inflammation in rheumatoid arthritis: an exploratory, multimodal biomarkers and neuroimaging study
Abstract
dc:description.abstractDepression is common in rheumatoid arthritis (RA) and is increasingly recognised as more than a psychological reaction to chronic disease. Evidence suggests a bidirectional relationship between RA and depression, with shared inflammatory pathways involving pro- and anti-inflammatory cytokines and chemokines. Peripheral inflammation can influence central nervous system (CNS) processes through multiple mechanisms, including altered blood–brain barrier function, microglial activation, and downstream effects on neuronal and astrocytic integrity. In parallel, neuroimaging provides in vivo markers of functional and structural brain phenotypes that may index inflammation-related changes relevant to mood and behaviour. However, despite the substantial clinical burden of depression in RA, there remains limited biomarker-driven understanding to support diagnosis, monitoring, stratification, or treatment targeting. This thesis applies an exploratory, multimodal approach to investigate neuroimmune correlates of RA-associated depressive symptoms by integrating serum CNS-derived proteins, peripheral inflammatory proteomics, and functional and structural MRI measures. Across analysed cohorts, RA participants exhibited higher circulating concentrations of CNS-derived proteins (neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), and total Tau) compared with healthy controls. Functional MRI connectivity analyses identified recurrent alterations implicating salience, default mode and fronto-limbic circuitry (including limbic–striatal and amygdala–prefrontal connections), observed in relation to CNS-derived proteins and, more modestly, mood symptom measures. Structural MRI yielded few robust associations, although amygdala grey matter volume showed opposite-direction relationships with depressive symptoms and Tumour Necrosis Factor (TNF)-α, highlighting limbic regions as priority targets for longitudinal investigation. Inflammatory proteomic profiling further identified a consistent association between fractalkine (CX3CL1) and both NfL and GFAP, alongside additional immune mediators, indicating a candidate CX3CL1-centred inflammatory signature linked to CNS stress markers. Nevertheless, while partial overlap across modalities was observed, the findings did not demonstrate a clear or robust association that jointly connected peripheral markers, neuroimaging phenotypes, and depressive symptom severity within RA. Overall, the results support the presence of measurable CNS-relevant biology in RA and highlight chemokine- and activation-skewed inflammatory pathways, together with salience/fronto-limbic network phenotypes, as promising candidates for further investigation. Given the apparent dissociation between individual biomarkers and depressive severity, the data favour a composite and longitudinal biomarker strategy that integrates serum CNS-derived proteins, selected inflammatory mediators, and connectivity metrics to characterise neuroimmune involvement and heterogeneity in RA. Larger, preregistered studies with enriched depression phenotypes and extended follow-up are warranted to validate these candidate signatures and assess their clinical utility for risk stratification, monitoring, and treatment targeting in RA-associated depression.
Degree
thesis:*- Level dc:type.qualificationlevel
- PhD
- Grantor dc:publisher.institution
- University of Glasgow
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Brock, James Loudoun
Subjects
dc:subject × 1Rights
- Language dc:language
- en
Identifiers
dc:identifier.*- DOI dc:identifier.doi
- https://doi.org/10.5525/gla.thesis.86182