Our project Systems Biology guided Machine Learning in Pharmacogenomics for Predicting Antidepressant Non-Remission In Late-Life Depression has just been granted by the DAAD.

The partner in the project is Prof. Daniel Mueller from the University of Toronto.

Abstract

Late-life depression (LLD), which is the occurrence of major depressive disorder (MDD) in older adults (age ≥60 years), is one of the most prevalent mental health disorders in later life. The presence of comorbidities, such as underlying cerebrovascular and neurodegenerative changes pose major challenges for diagnostics and treatment management, as LDD is pathophysiologically distinct from early-onset MDD. In combination with other challenges such as polypharmacy, more than 50% of patients experience relapse or do not achieve remission. In turn, persisting LLD symptomatology is associated with progressive cognitive decline and increased risk for dementia and stroke. While several mechanisms such as disruptions in subcortical, hippocampal, and frontotemporal regions have been proposed to play an important role for the comorbidities of LLD, evidence from genomics also suggests shared molecular pathways. Several genes associated with LLD play key roles in neurotransmitter signalling, neuronal plasticity and survival, as well as vascular processes. This genetic susceptibility underlying LLD and its comorbidities can be leveraged to predict antidepressant non-remission.