Prediction and structural equation model of sertraline treatment response in Japanese patients with major depressive disorder

Abstract

The aim of this study was to extract the factors possibly associated with sertraline treatment response and elucidate their interactions and extent of influence. Demographic state, stress state, personality, and eight genetic polymorphisms at baseline and clinical symptoms at baseline and 8 weeks were analyzed and examined by logistic regression and a structural equation model in sertraline treatment study of 96 Japanese patients with major depressive disorder. Non-responders were associated with higher scores of harm avoidance in Temperament and Character Inventory, higher scores (≥24) of 17-item Hamilton Rating Scale for Depression at baseline, recurrence, and 12/12 genotype of the serotonin transporter variable number of tandem repeat polymorphism in intron 2 (5HTTSTin2). When we calculated the response index using four factors extracted, the mean response index value of non-responders was significantly higher than that of responders. The symptoms at baseline, personality, recurrence, and polymorphism of 5HTTSTin2 showed significantly direct and positive influences on the symptoms at 8 weeks in our final structural equation model with a good model fit. Considering the combination of four factors extracted may be useful for predicting a worse response to sertraline treatment and selecting different treatment other than sertraline.