PM.20.003 – Biomarkers for personalizing treatment of depression

Route: Personalised medicine: the individual at the centre

Cluster question: 095 How can we personalise health care, for example by using biomarkers?

Depression is one of the leading causes of disability. Treatment of depression is based on stepped care, in which different treatment options are provided on a trial-and-error basis. Less than half of the patients benefit from each treatment step, leading to prolonged treatment trajectories that cause significant burden for patients and increase the costs for society. Recent studies have shown that treatment outcome can be predicted for individual patients using machine learning analysis of neuroimaging data (Cohen et al., 2020). This NWA project aims to develop biomarkers for treatment selection and validate those in a clinical trial. We will obtain clinical, neuroimaging, and blood markers of patients with major depressive disorder before their treatment. For biomarker development, we will recruit patients that are indicated for several of the current treatment steps, including treatment with psychotherapy, pharmacotherapy and neurostimulation. We will set up a Dutch consortium to enable the recruitment of patients and collection of data, measure treatment success using standard clinical criteria, and use state-of-the-art machine learning techniques for the development of biomarkers that predict treatment outcome for each of the treatment options. We aim to develop biomarker-based treatment algorithms in collaboration with stakeholders, and validate the superiority of biomarker-based treatment in a clinical trial. Thereby, this project aims to deliver a tool to aid clinical decision making for depression, which will shorten the treatment trajectory for patients, reduce the costs for society, and that can be readily implemented at other treatment centers.


Depression, machine learning, Neuroimaging, therapy


Organisation Amsterdam UMC (AMC)
Name Prof. G.A. (Guido) van Wingen