Inflammation and Oxidative stress in Early Psychosis: a multimodal data science and brain imaging approach

3rd supervisor: Professor Peter Liddle (University of Nottingham)

Activation of the innate immune system is of increasing interest in the pathogenesis of psychosis. Increased circulating IL-6 concentrations are present prior to psychosis onset and may be related to poor outcomes. Brain relevant local IL-6 inflammatory mediators include nitric oxide and chemokines with increased oxidative stress by generation of reactive oxygen and nitrogen species. A reduced defence to oxidative stress may potentially be responsible for brain changes seen in psychosis.

This PhD programme aims to use existing large data and newly acquired 1H- MRS in subjects with early psychosis, chronic schizophrenia and healthy controls to examine the clinical and neuroanatomical relevance and relationship between inflammatory state, oxidative stress and defence in the early stages of illness to test a number of developing hypotheses including eg 1) In psychosis, peripheral inflammation will be related to deficient central oxidative defences as measured by 1H-MRS 2) The deficient central oxidative defences mediate the effects of inflammation and will be related to structural brain changes (e.g., grey matter volume, cortical thickness and gyrification) and clinical symptom profiles (particularly impoverishment and disorganisation symptoms).

Existing data include that from the MRC funded SPRING (C.I. Liddle) and will build on the MRC funded PIMS (C.I. Upthegrove). Methods will include the modelling of symptom profile and brain structural changes associated with elevated peripheral markers of inflammation (IL6, TNFα) and oxidative defense (glutathione). The developed models will compare strength of immune signal effect size within differing clinical profiles with focus on negative symptoms and disorganisation of thought together with structural brain imaging (e.g., grey matter volume, cortial thickness and gyrification). Overall, methods will include a combination of classical parametric and non-parametric modelling (e.g., latent class analysis, linear mixed effect modelling) together with supervised and semi-supervised machine learning and other approaches from AI.

The student will also collect and analyse new primary data collected as part of the PIMS experimental medicine study of tocilizumab, an IL6 monoclonal antibody, in participants with early psychosis. This is a randomized, parallel arm, double-blind, placebo-controlled, single- dose experiment with an infusion of tocilizumab or normal saline given to participants with early psychosis. Data collected will include 1H-MRS assessments of glutathione before and after tocilizumab measured in the posterior medial prefrontal cortex and related areas and peripheral measurement of IL6, TNFα and glutathione. Finally, comparison of peripheral and 1H-MRS glutathione measurements and related structural MRI between early psychosis participants recruited in PIMS and chronic psychosis sample from SPRING will allow further exploration of the relevance of stage of illness

Person Specification
Applicants should have a strong background in psychology or psychiatry, and ideally a background in neuroimaging analysis (eg SPM, MatLab, Freesurfer) and some experience of data science. They should have a commitment to research in mental health and hold or realistically expect to obtain at least an Upper Second Class Honours Degree in a relevant subject.