Using early changes in cold cognition to predict response to vortioxetine in major depressive disorder

Abstract

Antidepressant pharmacotherapy dominates current treatment in psychiatry, including treatment for major depressive disorder (MDD). However, the current trial-and-error process of medication selection contributes to treatment failure and unnecessarily exposes patients to lengthy and insufficient treatment trials. Notably, improvements in measures of cognition have been demonstrated to occur early during treatment and prior to improvements in clinical state. Cognitions have been categorized based on emotional valence (i.e., cold versus hot cognitions). Cold cognitions describe cognitive operations that are relevant to the processing of non-emotional information. The current analysis investigates whether early changes in cold cognition can predict response after 8 weeks of vortioxetine treatment in adults with MDD. This was secondary analysis of an 8-week, open-label study. Cognition was assessed at week 0 and week 2 to measure early cognitive change. Depressive symptom severity was assessed at week 0 and week 8 to measure treatment response. Eighty-one subjects were analyzed using binomial logistic regression models. Early change in cognition was a non-significant predictor of response (p = 0.845, SE = 0.599, OR = 1.124), which may have resulted from high data variability. The overall predictive accuracy of the model was low (sensitivity = 37.5%, specificity = 89.8%, PPV = 70.6%, NPV = 68.8%). Future studies should include larger samples and stratify patients based on potentially moderating variables, such as baseline cognitive impairment and occupation. Stratification would likely produce more homogenous samples, reducing the amount of variability observed for early cognitive change.