Introduction: Despite various marketed antidepressants, non-response is still a major problem in depression therapy. One way of enhancing therapy outcomes could be to apply biomarkers predictive for response. Additionally, susceptibility biomarkers might be used in depression prevention, development of new therapeutic approaches or understanding the disease.

Aim: This master’s thesis evaluated genetic markers for response and susceptibility, as well as the use of the cordance as predictor for response.

Methods: 37 patients experiencing depressive episodes were randomly assigned to either a control (arm A) or an intervention group (arm B) and observed for 35 days. Participants received a new antidepressant according to doctor’s choice on day 1. After one week of treatment, a sleep EEG was performed. In arm B, the therapy was adapted in case of negative cordance prediction. The primary endpoint was the responder rate on day 35, assessed using HAMD-21-scores. Genetic markers were analysed using genotyping.

Results: No significant effects of therapy adaptation based on cordance prediction were observed. The rs1360780 wildtype and the CYP2D6 IM phenotype were significantly more frequent in responders, whereas the CYP2C19 UM phenotype was more frequent in non-responders. There was a trend for the rs1065852 being more frequent in responders compared to non-responders. Significant differences in allele or genotype distribution between the study population and a reference population were observed for rs2032583, rs12248560, rs6277, rs463376, rs36029, rs192303, rs774676466, MAOA u-VNTR, DRD4 exon 3 VNTR and CYP2C19 UM phenotype.

Conclusion: Interesting associations were observed. Larger studies are necessary to prove their predictive power.