This is the first study to investigate the interaction between temperament traits, antidepressant treatment response and
TPH1 A218C genotype. Associations between
TPH1 genotype and temperament dimensions were studied in patients with major depressive disorder and linear changes were found at both baseline and endpoint. Subjects with the C-allele of A218C scored higher on HA and lower on NS. HA scores were higher at baseline than at endpoint and NS scores increased from baseline to endpoint. In multivariate analysis of covariance (MANCOVA)
TPH1 genotype was used as a factor and MADRS scores as a covariate, since depressive symptoms have been found to modify temperamental traits and, on the other hand, temperament has an impact on recovery from depression [
18]. HA and NS scores at baseline were analyzed as temperament profile itself can reflect the biological subtype of depression and be associated with the clinical response. Treatment response has been found to affect HA and NS scores at endpoint [
20]. A possible interaction of
TPH1 genotype and temperament dimensions was not separately analyzed in treatment response due to the obvious associations detected between HA and NS dimensions and
TPH1 genotypes. Additionally, an interaction of remission status and most markedly with
TPH1 CC genotype was found to be associated with endpoint HA score. According to this analysis, the endpoint HA score variation was greater with the CC genotype than with the AG/AA genotypes when the remission and non-remission groups were compared. This finding may indicate the CC genotype being associated more with traits related to risk of depression, including the tendency to harm avoidance, than with the depressive state itself. Earlier studies, including one meta-analysis, have reported contradictory results on the relationship between
TPH1 genotype and treatment response in major depression [
2,
4]. In a population based study no association was found between
TPH1 genotype and TCI temperament dimensions [
21]. Both the depressive disorder among our patients and the different ethnic background may explain the discrepancy in the results. It was therefore important to study the possible interactive effect of genotype and treatment response on temperament dimensions. As the impact of genotype alone is likely very small it can be hypothesized to be connected with the depressive trait rather than the clinical state.
These results suggest that C allele of
TPH1 is associated with differences in temperament profile. High HA among C allele carriers may lead to general avoidance behavior and susceptibility to depression. There is one earlier study reporting no association between temperament dimensions and
TPH1 genotype [
5], but in this sample most of the patients had a diagnosis of bipolar disorder. The different findings likely reflect the different genetic backgrounds of bipolar and unipolar depressive disorders.