A Phase 2 randomised, double-blind, placebo-controlled trial of emestedastat in patients with major depressive disorder and cognitive impairment.

BACKGROUND

Most antidepressants do not independently treat impaired cognition associated with major depressive disorder (MDD). Elevated brain cortisol levels are associated with both MDD and cognitive impairment. Emestedastat is a brain-penetrant, intracellular cortisol synthesis inhibitor.

AIMS: The XanaCIDD trial aimed to determine whether emestedastat 10 mg could improve both cognitive impairment and depression in a population with persistent MDD and cognitive impairment.

METHOD: Participants had a diagnosis of MDD, with persistent depression (Hamilton Rating Scale for Depression-17 ≥17) and cognitive impairment (coding test ≤0.5 s.d. of normal). Randomised treatment (1:1) was continued for 6 weeks (week 6), with 4 weeks' blinded follow-up (week 10). The primary end-point was cognition (composite of attention/working memory) at week 6. Depression end-points included Montgomery-Åsberg Depression Rating Scale (MADRS) and Patient Global Impression of Severity (PGI-S). Analyses used a mixed model for repeated measures (cognitive impairment and MADRS one-sided tests, others two-sided) and Cohen's d effect size. No adjustments were made for multiple comparisons.

RESULTS

A total of 165 participants were enrolled and treated: 134 (81%) were on background antidepressants, 62% were female, mean prior number of MDD episodes was 10 and mean coding z-score was -1.47. There was no benefit on the primary cognitive end-point at week 6 (p = 0.17 favouring placebo), with the cognitive composite improving markedly in both groups without correlation to depressive symptoms or baseline characteristics (R2 ≤ 0.02). The trend towards benefit on the prespecified secondary depression end-point, MADRS, began at week 6 and was maximal at week 10 (2.7 points, Cohen's d 0.43, p = 0.05). A trend towards potential emestedastat benefit in PGI-S scores was also maximal at week 10 (p = 0.12). Mild-moderate transaminase elevations were seen in 10% of emestedastat participants (≤2 times normal in 7 of 8 cases, 1 discontinuation) versus 1% of the placebo group.

CONCLUSIONS

There was no benefit of emestedastat on the primary end-point of cognitive impairment and a large placebo effect, without correlation of cognitive impairment to depression changes or baseline characteristics. Trends towards potential antidepressant activity suggest that emestedastat may be a novel antidepressant worthy of further investigation.