GENOME WIDE ASSOCIATION STUDY OF TREATMENT RESPONSE TO COGNITIVE BEHAVIORAL THERAPY FOR DEPRESSION

Background Major Depressive Disorder (MDD) is a serious, but treatable, disorder where Cognitive Behavioral Therapy (CBT) is a first line treatment. However, a significant number of patients (40–60%) are not responding to treatment and understanding more about the disorder and who responds to treatment is crucial to improve outcomes. In the past year, there have been marked advances in the genomics of MDD thanks to large consortia GWAS efforts. Here we leverage summary statistics derived from the latest MDD GWAS in order to test whether individual MDD polygenic risk scores (PRS) associate with baseline severity and CBT response in a sample of 971 MDD cases undergoing internet-based CBT (iCBT). Our aim is to calculate MDD Polygenic Risk Scores (PRS) and examine their relationship with a key measure of disease severity and treatment response with longitudinal follow up data. Methods We collected DNA from 971 Swedish MDD cases (66% female) who took part in a 12-week course of iCBT. The primary outcome variable was the Montgomery Asberg Depression Rating Scale – Self-rating (MADRS-S) and was measured pre-treatment, weekly, post treatment and at 6 months follow-up. For genotyping, the Illumina Global Screening Array (GSA) was used for the 971 samples. For population-specific controls, we will calculate MDD PRS for a set of 3,000 Swedish individuals with no history of MDD. We use the latest MDD summary statistics to calculate PRS for each individual to examine the relationship between polygenic load and a key measure of disease severity and treatment response. Further, we calculate PRS for a range of other traits which may influence disease severity or treatment response: neuroticism, bipolar disorder, subjective well-being and educational attainment. Results We will present results detailing the relationship between MDD PRS and disease severity (baseline MADRS) as well as treatment response (change in MADRS following iCBT) and treatment adherence. Further, we will present results demonstrating how PRS for a range of other traits influence disease severity or treatment response. Discussion In other psychiatric disorders, such as schizophrenia, polygenic risk score analyses and treatment response of pharmacological interventions have been investigated with interesting outcomes. In our future work, we will contrast genetic predictors with non-genetic predictors derived from the Swedish registers (e.g. pre- and perinatal risk factors), to try to understand even more about the biological underpinnings of MDD and response to treatment. We believe performing these analyses will bring us one step closer to determining whether genomics can help personalize MDD treatment.