STRATIFY - ESTRA Follow-Up Study
Research type
Research Study
Full title
Brain network based stratification of reinforcement-related disorders - ESTRA (Eating Disorders) Follow-up Study
IRAS ID
272117
Contact name
Sylvane Desrivières
Contact email
Sponsor organisation
Kings College London
Clinicaltrials.gov Identifier
218030, STRATIFY IRAS No.
Duration of Study in the UK
4 years, 9 months, 22 days
Research summary
We are interested in understanding the connections between certain behaviours such as negative emotions, difficulties with attention, and unusual experiences. In this project we will investigate biological and social causes of Eating Disorder. We expect that the results of this study will develop better treatments for people who have Eating Disorders and related mental health difficulties, by tailoring treatments to precisely target those symptoms that cause suffering in each individual patient.
We will use functional and structural neuroimaging, cognitive, emotional as well as environmental data to understand if there are sub-groups within the ED diagnoses that we will investigate, and also if there are similar processes across these ED diagnoses. We will also identify and characterize predictors of future psychopathology.
To do so, we will follow-up 100 participants with Eating Disorders (anorexia nervosa and bulimia nervosa) and 50 healthy controls who have previously participated in the STRATIFY study. In addition, we will recruit 100 new participants who have been recovered from an Eating Disorder (anorexia nervosa or bulimia nervosa) for over 6 months (defined by the SCID-5 interview). We will use advanced statistical methods using data such as genetics, neuroimaging, behavioural measures that allow us to make predictions for how symptoms of eating disorders might change. The clusters of traits that we will identify (genetic, neuroimaging, behavioural) resulting from this analysis will be compared to those previously identified in a longitudinal population-based genetic and imaging study, the IMAGEN sample of over 2000 participants spanning the period from adolescence to adulthood. Identifying specific neural processes in both this clinical sample (STRATIFY – ESTRA) and also the population-based sample (IMAGEN) will enable us to develop targeted early intervention strategies and also identify of novel pharmaceutical targets.
The research will take place at relevant NHS sites and research centres in London and participants will undergo two research visits, including one brain scan, a blood sample and a range of social and behavioural measures.
Summary of Results
The ESTRA cohort has been utilised extensively to identify predictors of disease onset, in hopes that these predictors can be used to detect individuals at risk of developing psychiatric disorders, to fast-track them for treatment. Using machine learning methods, Desrivières et al. (2024), demonstrated that it was possible to distinguish patients with eating disorders (EDs), Major depressive disorder (MDD), and Alcohol use disorder (AUD) from healthy controls (HC), using psychosocial data, including personality, cognitive and environmental questionnaires. Neuroticism, hopelessness and ADHD symptoms were shred features that contributed to the accurate classification of the ED, MDD, and AUD groups from healthy controls, highlighting a shared contribution of these factors to general mental health disorder development. Zhang et al. (2024) explored the association between personality and ED development in more detail and demonstrated that hopelessness and neuroticism were key personality characteristics that distinguished healthy controls and ED patients.Several studies have used the ESTRA cohort to identify neurobiological differences between patient groups and age-matched controls, which may serve as targets for future treatment development. The study by Yu et al. (under review) identified brain alterations in Anorexia Nervosa (AN) and Bulimia Nervosa (BN) patients, which were associated with cognitive restraint, alongside lower activation during reward processing. Alterations in reward processing were mediated by neuroticism rather than cognitive restraint, indicating the role of personality traits in neural responses. The study by Xiang et al. (under review) uses ESTRA data to explore how the amygdala contributes to processing emotions in facial expressions. Researchers identified separate processes for emotional evaluation and uncertainty in the amygdala during tasks involving emotional faces. Additionally, a study by Lett et al. (under review) uses data from IMAGEN, STRATIFY, and ESTRA to explore the shared clinical and neurobiological aspects of mental health disorders. Researchers integrated data from clinical symptoms and brain imaging in young adults and identified six distinct patterns of psychopathology (mania, depression, anxiety, stress, eating habits, and panic attacks) linked to various brain features. These patterns were consistent across different age groups and psychiatric patient samples, suggesting associations between symptoms and brain function.
By using data from the ESTRA study, Xie et al. (2023) identified a common mechanism that underlies multiple mental health disorders’ symptoms. Delayed development of the brain’s frontal regions was identified to be the common cause of these mental health disorders. Frontal regions such as the Prefrontal Cortex are responsible for decision-making and planning processes. In turn, they found that this common factor, which is partially genetically determined, can be identified from pre-adolescents to young adulthood. The identification of this common factor acts as a 'target' used to develop early intervention strategies and future therapies.In a recently published study assessing mental health trajectories across the covid-19 pandemic. Qi et al. (2023) found that mental health status pre-pandemic predicted changes in symptom severity across the pandemic. Interestingly, compared to the population cohort (IMAGEN), participants with depression and eating disorders from the ESTRA and STRATIFY cohorts reported a lasting decrease in depressive and eating disorder symptoms.
King et al. (under review) investigated the link between childhood trauma (CT) and cognitive impairment across different psychiatric disorders using ESTRA data. They found that reduced connectivity in the brain default mode network (DMN) during social cognitive tasks was associated with higher CT scores and worse cognitive performance. The DMN is a network of different brain regions that is active in states such as wakeful rest, daydreaming and mind-wandering. The findings suggest that abnormal DMN connectivity may play a role in impaired emotional regulation and cognitive function across psychiatric disorders, with stronger effects observed in depression
REC name
North West - Greater Manchester South Research Ethics Committee
REC reference
20/NW/0143
Date of REC Opinion
17 Mar 2020
REC opinion
Favourable Opinion