GENETICALLY DEFINED SUBGROUP ANALYSIS REVEALS DIFFERENTIAL RESPONSE TO VASOPRESSIN V1B ANTAGONISM IN PATIENTS WITH MDD PRE-EXPOSED TO ANTIDEPRESSANT IN THE CURRENT EPISODE: POST-HOC ANALYSIS OF THE PHASE 2 OLIVE TRIAL

Background

Treatment-resistant depression (TRD), a subtype of Major Depressive Disorder (MDD) defined by suboptimal response to antidepressant treatment, represents a significant unmet medical need. Vasopressin V1b receptor modulation has emerged as a potential mechanism relevant to stress-system dysregulation in MDD. BH-200, a selective V1b receptor antagonist, was evaluated in a Phase II randomized, double-blind, placebo-controlled trial (OLIVE, CTIS: 2024-513104-34-00) in adults with MDD. The estimated mean improvement from baseline to Week 8 in the mITT population (N=331) was larger in the BH200-treated arm than in the placebo arm (∆ = -2.98, p = 0.0003). To evaluate BH-200’s potential for use in TRD, a post-hoc subgroup analysis was performed in OLIVE’s MDD patients who had not responded to at least one adequate course of antidepressant treatment in their current depressive episode prior to enrolling into the trial. This subgroup is defined here as pre-exposed MDD. To test whether variation in V1b-related biology identifies likely responders, we applied a refined polygenic classification retrospectively to the pre-exposed MDD subjects.

Objectives

This analysis aimed to determine the antidepressant efficacy of BH-200 compared with placebo in pre-exposed MDD participants and evaluate whether the polygenic classification is associated with efficacy of BH-200, in the pre-exposed MDD cohort. In addition, we aimed to assess the consistency of findings across HAM-D17 and MADRS outcomes.

Methods

Data were obtained from the randomized, double-blind, placebo-controlled OLIVE trial (mITT N=331), including a pre-exposed MDD subset defined by non-response to at least one antidepressant treatment (n=171). The target endpoint for this analysis was HAM-D17 change from baseline to Week 8. Mixed-model repeated measures (MMRM) analyses were applied. The polygenic classification tool, trained post-hoc using repeated nested cross-validation in the OLIVE dataset to predict BH-200 treatment response, was applied to compute test-positive and test-negative strata.

Results

A total of 171 pre–exposed MDD participants were included. BH-200 showed greater improvement versus placebo in LS mean change from baseline on HAM-D17 in pre-exposed MDD cohort. The treatment difference at Week 6 was -2.00 (p=0.029), and at Week 8 it was 1.45 (p=0.17). In the genetically defined test-positives (n=46), at Week 6, LS mean change over placebo was −5.65 (p=0.00040), and at Week 8, was –6.03 (p=0.00079). Findings were broadly consistent across HAM-D17 and MADRS outcomes. Discussion and

Conclusion

BH-200 demonstrated a clinically meaningful signal in preexposed MDD participants without application of genetic classification. Use of the polygenic classification tool improved signal detection in the test-positive group underlining its potential utility for precision psychiatry approaches in future BH-200 studies. Consistent findings across HAM-D17 and MADRS strengthen confidence in the clinical relevance of V1b pathway modulation for the treatment of depression. Further trials using biomarker-based enrichment are warranted.