Predictive neural biomarkers of clinical response in depression: a meta-analysis of functional and structural neuroimaging studies of pharmacological and psychological therapies

We performed a systematic review and meta-analysis of neural predictors of response to the most commonly used, evidence based treatments in clinical practice, namely pharmacological and psychological therapies. Investigations of medication-free subjects suffering from a current major depressive episode who underwent positron emission tomography (PET) or functional or structural magnetic resonance imaging (MRI) scans prior to the initiation of treatment were reviewed. Results of 20 studies from 15 independent samples were included in the functional imaging meta-analysis and 9 studies from 6 independent samples in the structural neuroimaging meta-analysis. Regional activations with prognostic value include the well replicated finding that increased baseline activity in the anterior cingulate is predictive of a higher likelihood of improvement. As well, increased baseline activation in the insula and striatum is associated with higher likelihood of a poorer clinical response. Structural neuroimaging studies indicated that a decrease in right hippocampal volume is a statistically significant predictor of poorer treatment response. Overall, the predictive information that is measurable with brain imaging techniques is both multimodal and regionally distributed as it contains functional as well as structural correlates which encompass several brain regions within a frontostriatal-limbic network. To develop clinically relevant, prognostic markers will require high predictive accuracy at the level of the individual. Predicting clinical response will help to stratify patients and to identify at an early stage those patients who may require more intensive or combined therapies. We propose that structural and functional neuroimaging show significant potential for the development of prognostic markers of clinical response in the treatment of depression.