From DNA to CBT : an investigation psychological therapy response and genetics

Abstract: Background: Social anxiety disorder, panic disorder, and major depressive disorder are common, disabling, and heritable disorders. The first line of treatment for all anxiety disorders and mild to moderate depression is cognitive behavior therapy (CBT). However, 40–60% of patients fail to respond adequately to treatment and the ability to a priori predict who will respond to treatment is essential for being able to choose alternative treatments to those who will fail the treatment. As clinical and demographic variables so far have shown limited predictive value of in guiding therapeutic decisions, is has been suggested to incorporate genetic variation in the set of predictors to increase precision. Aims: The general aim of this thesis was to investigate the genetic underpinnings of psychological treatment outcomes for social anxiety disorder, panic disorder, and major depressive disorder. Specifically, we aimed to identify the clinical and genetic predictors of patients with social anxiety, and to evaluate participant responses to CBT (Study I); next we increased the sample size from Study I and collaborated with another site and investigated three polymorphisms and their predictive value in CBT response for social anxiety (Study II). Subsequently, we aimed to calculate polygenic risk scores by performing a genomewide association analysis to study genetic variation and CBT responses to major depression (Study III); finally, we further extended the sample size by performing a meta-analysis, and by calculating polygenic risk scores to study potential genetic overlap with other cognitive and psychiatric traits, and the link between genetic variations and CBT responses in patients with anxiety disorders. Methods: In Study I and II, we recruited participants from randomized controlled trials to investigate clinical variables (Study I) and candidate gene polymorphisms (Study I and II); as potential predictors of CBT response for social anxiety disorder. Study III and IV we recruited larger samples and performed genome-wide association analyses to estimate genetic variance in CBT response and to calculate polygenic risk for CBT outcome for major depression and anxiety disorders. Results: In Study I, several clinical, but neither of the genetic variables predicted CBT outcome for social anxiety. In Study II, neither of the genetic polymorphisms predicted response to CBT for SAD. In Study III, an association of higher load of autism spectrum polygenic risk scores and less decrease in depressive symptoms after CBT was shown. In Study IV, no genome-wide significant loci or genetic overlap with psychiatric or cognitive traits were present. Conclusions: The results in this thesis presents the first findings of an association of aggregated genetic risk scores and CBT outcome in depression. Our results indicate that the use of polygenic risk scores may be a fruitful approach in genetic studies of CBT outcomes. In addition, the results provide support for the continued efforts of collecting larger and more homogenous samples in studies of complex traits such as psychological treatment response.