A Randomized Double-Blind Trial of D-Cycloserine Augmented Single-Session Exposure Therapy for Anxiety: Neurocognitive Mechanisms

Abstract

Background - Drugs targeting the N-Methyl-D-aspartic acid (NMDA) system and learning rates have been proposed as potential adjunct treatments to boost the success of exposure therapy for anxiety disorders. However, the effects of the NMDA partial agonist d-cycloserine on psychological treatment have been mixed. We therefore investigated potential neurocognitive mechanisms underlying the clinical effects of d-cycloserine-augmented exposure therapy using a single-session testbed, to inform the optimal combination of this and similar agents with psychological treatment. Methods - This randomized double-blind study was run at a single location (university environment). Unmedicated patients with panic disorder received single-dose d-cycloserine (250mg; N=17) or matching placebo (N=16) 2hrs before one session of exposure therapy. Clinical symptom severity was measured using self-report and clinician-rated scales the day before and after treatment, and at 1- and 6-months follow-up. Neurocognitive markers were assessed one day after treatment, including two primary outcome measures threat bias for fearful faces (FDOT) and implicit panic associations (IAT) and a secondary outcome measure amygdala response to threat. Analysis was by intention-to-treat. Outcome - Threat bias (FDOT) reduced to a larger extent following d-cycloserine (CYC) versus placebo (PLAC) but failed to reach the corrected level of significance (p < 0·025) (mean difference -19·0, 95% CI -38·04 to -0·14; p=0·042, d=0·77). There were no significant drug effects on the IAT (mean difference -0·01, 95% CI -0·32 to 0·29; p=0·93, d=0·11). On the day after treatment, amygdala threat response was lower in the drug compared to the placebo group (mean difference -0·32, 95% CI -0·61 to -0·03; p=0·023, d=1·08), with lower magnitude predicting greater clinical improvement during follow-up across groups. D-cycloserine led to greater clinical recovery at 1-month follow-up (CYC 71% versus PLAC 25%, RR 2·55, CI 1·16-5·61, p=0·015). No serious side effects were reported. Interpretation - D-cycloserine-augmented single-session exposure therapy reduces amygdala response to threat, and this effect predicts later clinical response. These findings highlight a neurocognitive mechanism by which d-cycloserine may exert its augmentative effects on psychological treatment and bring forward a marker that may help understand and facilitate future development of adjunct treatments with CBT for anxiety disorders. Trial Registration Number: clinicaltrials.gov, NCT01680107 Funding - Medical Research Council, Oxford University John-Fell-Fund (D-cycloserine Augmented CBT for Panic Disorder; clinicaltrials.gov; NCT01680107). Declaration of Interest: AR and AN report no conflicts of interest. MB has received travel expenses from Lundbeck for attending conferences and has acted as a consultant for J&J. CH has received consultancy fees from Lundbeck, P1vital, J&J, Pfizer and Servier. Ethical Approval: Ethical approval was obtained from the South Central - Hampshire research ethics committee. All participants gave written informed consent.