Developing a neurocognitive model of the antidepressant effects of ketamine

Background: Major depressive disorder (MDD) affects approximately 1 in 10 people, and is the second leading cause of disability worldwide (Whiteford et al., 2013). Although 67% of patients with MDD respond to antidepressant medication after one or more acute treatment steps (Rush et al., 2006), approximately 30% have illness resistant to treatment (Rush et al., 2006; Ionescu et al., 2015). Treatment-resistant depression (TRD) costs the UK government up to an additional £25,000 per person per year compared to non-TRD (McCrone et al., 2017), as the search for therapies that are effective, easily administered and well-tolerated by patients continues.

Recent research suggests that the N-methyl-D-aspartate (NMDA) antagonist ketamine can be effective in TRD (Murrough et al., 2013). Indeed, compelling evidence has led to the licensing of the ketamine metabolite esketamine by the Food and Drug Administration (FDA) for use in conjunction with an oral antidepressant in TRD in the United States. In contrast to conventional antidepressants such as selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs), which take up to six weeks to show a therapeutic effect, the antidepressant effects of ketamine are observed within hours of dosing (Zarate et al., 2006). Preclinical animal models suggest these timelines are a reflection of the drugs' distinct mechanisms of action. A study by Stuart et al. (2015) on contextual affective learning in rodents showed that the SNRI venlafaxine promoted the acquisition of positive bias (preference for SNRI-paired rewarded digging substrate) when administered before learning, while ketamine reduced negative bias (aversion to stress-paired rewarded digging substrate) associated with prior aversive learning. These findings suggest that ketamine reduces the retrieval of established negative memories, and that the timescale for the onset of the antidepressant effects of ketamine is shorter than that of conventional antidepressants due to it being less dependent on environmental exposure for learning and relearning positive cue-outcome associations.
It is suggested that ketamine may exert its clinical effects via the lateral habenula (LHb), an area implicated in predicting negative events and responding to unexpected negative outcomes (Matsumoto et al., 2007, 2009). In support of this, the burst firing of LHb neurons has been shown to be supressed by ketamine injection one hour before recording in a rodent model of depression (Yang et al., 2018). Data from human functional magnetic resonance imaging (fMRI) studies support a role for abnormal LHb activity in learning about negative outcomes in depression, as this activity was diminished during the prediction or experience of monetary punishment in individuals with MDD but not in healthy controls (Furman & Gotlib, 2016).
In humans, modern experimental medicine studies have shown that conventional antidepressants reverse negative affective biases on tests of emotional processing in depressed individuals, and that these effects arise well before detectable changes in mood (Harmer et al., 2009; Harmer et al., 2017). Such findings support animal literature suggesting that conventional antidepressants promote more positive associative learning, where a certain length of exposure to environmental stimuli is necessary for progression into clinical response. Nevertheless, the pre-clinical findings on ketamine have yet to be translated to clinical research, such that the neuropsychological mechanisms by which ketamine produces improvements in mood in patient populations remain relatively unexplored. This project aims to develop a neurocognitive model of fast-acting antidepressant drug efficacy in humans, with a particular focus on the effects of ketamine on emotional processing, learning and memory.