Prediction of response to antidepressant monotherapy

The Munich Antidepressant Response Signature (MARS) Project aims to find predictors for antidepressant response. Eighty-one patients were treated either with mirtazapine or paroxetine for at least 4 weeks (starting within the first 3 weeks after admission). Forty patients were classified as remitters/responders and 41 as non-responders. Remitters/responders did not significantly differ from non-responders in diagnoses, psychopathology, family history, number of previous episodes, the Dex/CRH test on admission, age, and gender. In the Dex/CRH test prior to discharge remitters/responders showed significant reductions in cortisol and ACTH responses compared to admission. However, non-responders showed increased cortisol and ACTH responses prior to discharge. These changes were independent of specific antidepressant treatment (mirtazapine, paroxetine). These results show that psychopathological and neuroendocrinological parameters alone are insufficient for the prediction of response to antidepressant monotherapy. The combination of psychopathology with endophenotypes and genetic polymorphisms will probably improve prediction of response to antidepressant treatment in the future.