MRI measures of brain structure and function as predictors of medium-term response to psychotherapy in adolescent MDD

Major depressive disorder (MDD) in adolescence is a challenging priority for NHS mental health services as well as being a difficult experience for the patient and their family. The recommended treatment is psychotherapy, which can be combined with antidepressants if there is no immediate response. Unfortunately, there is no way currently to predict which patients are most likely to respond to psychotherapy. This uncertainty stems in part from our incomplete knowledge of how depression is caused by abnormal function of brain circuits important for mood states and their conscious control. One plausible theory we want to investigate is that depressive symptoms result from an overactive response to negative events or stimuli in one part of the brain (the limbic system), which is not effectively controlled by an underactive response in a different but interconnected brain region (the frontal cortex). We predict that this functional imbalance in fronto-limbic brain networks can be normalised following cognitive behavioural therapy, a form of psychotherapy focused on improving conscious control of mood symptoms. We also expect that the extent of structural and functional abnormality measured in fronto-limbic brain networks before treatment begins can predict patients? responses to treatment up to one and a half years later.

To test these predictions, we are proposing to collect magnetic resonance imaging (MRI) measurements of brain structure and function in 120 adolescent patients with MDD and 40 health adolescent volunteers. All of the patients will be scanned once before they are randomly assigned to one of three different kinds of psychological treatment as part of a larger study, already funded to investigate the clinical effectiveness of psychotherapy for young people with depression. Forty of the patients (and the healthy volunteers) will also be scanned on a second occasion 20 weeks later. MRI scanning is a painless and safe method of obtaining detailed information on the structure and function of the human brain. The design and clinical governance of this study has been informed by a careful consideration of the risks and benefits to adolescent patients and volunteers. We consider that it is timely and important to use modern brain scanning techniques to better understand and predict the course of this serious mental illness in young people and that this research proposal will safely and powerfully test key theoretical predictions in a way that could improve the future clinical care of adolescent patients with major depressive disorder.

Adolescent major depressive disorder (MDD) is a challenging clinical condition associated with considerable morbidity and mortality as well as poor outcome in adult life, and has been relatively under-investigated by neuroimaging studies to date. Symptomatic response to clinically recommended psychological treatments is variable, but little is known about the neurobiological basis of this variability in adolescent patients and there are few tools available to predict therapeutic response. This proposal will evaluate candidate neuroimaging markers and predictors of response to psychological treatment of adolescents with MDD in the context of a larger-scale, NIHR-funded study of symptomatic and social outcomes measured up to 1.5 years after treatment.

Consistent with existing data on adult MDD, we hypothesise i) that depressive disorder in adolescents will be associated with abnormalities of brain function in a fronto-limbic brain system including the amygdala, anterior cingulate cortex and prefrontal cortex; ii) that these abnormalities at baseline assessment will be ?normalized? by cognitive behavioural therapy; and iii) that baseline measures of fronto-limbic brain structure and function will be predictive of medium-term clinical and social response to psychological treatments.

These hypotheses will be tested by acquiring baseline structural and functional MRI data, immediately prior to initiation of treatment, from a cohort of 120 adolescent MDD patients randomised to one of three psychological treatments: cognitive behavioural therapy (CBT), brief psychodynamic psychotherapy or active clinical care, as part of the IMPACT clinical trial. The subset of 40 patients randomised to receive CBT will be scanned on a second occasion 20 weeks later, after completion of psychotherapy. Functional MRI data will be acquired during performance of emotion-processing and emotionally valent executive and memory tasks, as well as while subjects are at rest. Brain structure will be measured by conventional MRI and diffusion tensor imaging. Image analysis will involve innovative approaches to assessment of endogenous dynamics and complex brain networks in addition to well-validated methods for non-parametric inference on more conventional activation and morphometric statistics. These imaging markers measured in patients at baseline and follow-up will be compared to the same markers measured in an age-matched group of healthy volunteers (N=40) to test hypotheses (i) and (ii); baseline markers on all patients will be used to test hypothesis (iii) and validate candidate neuroimaging predictors of treatment response.