NEUROACTIVE STEROID CHANGES PRE-AND POST-KETAMINE DURING A DOUBLE-BLIND, CROSSOVER CLINICAL TRIAL: AN EXPLORATORY STUDY

Background

Neuroactive steroids such as allopregnanolone, pregnenolone, and dehydroepiandrosterone (DHEA) modulate GABAergic and glutamatergic signaling and are implicated in mood regulation and rapid-acting antidepressant mechanisms. Although effective in treatment-resistant depression (TRD), ketamine’s effects on endogenous neurosteroid dynamics remain underexplored. Conversely, how neurosteroid dynamics contribute to ketamine’s antidepressant mechanisms also remains unknown.

Methods

In this double-blind, placebo-controlled, crossover study, 32 unmedicated adults with TRD received a single intravenous infusion of ketamine (0.5 mg/kg) or saline placebo, with crossover after a seven-day washout. Plasma concentrations of allopregnanolone, pregnenolone, DHEA, cortisol, adrenocorticotropic hormone (ACTH), and brain-derived neurotrophic factor (BDNF) were measured at four timepoints relative to infusion: –60 min, +230 min, +24 hr, and +72 hr. Depression severity was assessed using the Montgomery-Åsberg Depression Rating Scale (MADRS). Linear mixed-effects models examined biomarker trajectories and associations with antidepressant response. A random intercept cross-lagged panel model (RI-CLPM) was applied to test temporal directionality and within-person dynamics among select biomarkers.

Results

Compared to placebo, pregnenolone levels significantly declined over time following ketamine infusion (p=0.043). No significant group-level changes were observed for allopregnanolone, DHEA, or other biomarkers. Higher cortisol levels at early post-infusion timepoints were associated with less reduction in MADRS scores over time (p=0.021), suggesting blunted antidepressant response. Neurosteroid levels were not significantly predictive of treatment outcome. Longitudinal modeling revealed that allopregnanolone levels remained relatively stable across timepoints within individuals, in contrast to other neurosteroids— suggesting it may reflect a more trait-like biological signature. Interestingly, higher baseline BDNF levels were significantly associated with lower post-ketamine allopregnanolone levels (β = –0.19, p = 0.004); this effect was observed only when adjusting for covariates. No directional associations emerged between pregnenolone and its downstream metabolites across time.

Conclusions

Ketamine selectively modulated upstream neurosteroidogenesis, particularly pregnenolone, without robust downstream changes. Allopregnanolone’s within-subject stability may indicate a trait biomarker rather than a dynamic state marker. The inverse association between BDNF and allopregnanolone highlights a potentially reciprocal relationship in neuroplasticity and neurosteroid signaling following ketamine. Given the overlapping therapeutic relevance of both ketamine and synthetic neurosteroids (e.g., in postpartum depression), the neuroactive steroid system represents a promising translational target for future mechanistic and clinical investigation.