Zuranolone efficacy and safety by baseline symptom severity on the Montgomery-Asberg Depression Rating Scale: post-hoc analysis of SKYLARK and ROBIN

Introduction

Postpartum depression (PPD) is a common perinatal condition with significant consequences for patients as well as their children and families. Zuranolone, an oral, once-daily, 14-day treatment for PPD, was evaluated in two phase 3 randomised, placebo-controlled trials for PPD: SKYLARK[1] and ROBIN[2]. In these trials, efficacy was evaluated using the Hamilton Depression Rating Scale 17 (HAMD-17) total score as a primary measure and the Montgomery-Asberg Depression Scale (MADRS) total score as a secondary measure. Both studies enrolled participants with severe baseline depressive symptoms according to baseline HAMD-17 total scores (total score ≥26). This post hoc analysis of SKYLARK and ROBIN characterised the efficacy and safety of zuranolone in patients with PPD with either moderate or severe depressive symptoms as defined by baseline MADRS total score.

Methods

For the purposes of this analysis, subgroups were defined as having moderate (MADRS total score, 20-34) or severe (MADRS total score, 35 or higher) baseline depressive symptoms. Changes from baseline in HAMD-17 and MADRS total scores were analysed for each subgroup using a mixed-effects model for repeated measures. Treatment-emergent adverse events were also evaluated by subgroup.

Results

Among 344 participants with PPD (SKYLARK, n=194; ROBIN, n=150), 43.6% (SKYLARK, n=89; ROBIN, n=61) met criteria for the moderate depressive symptoms subgroup and 56.4% (SKYLARK, n=105; ROBIN, n=89) met criteria for the severe depressive symptoms subgroup based on baseline MADRS total scores. Demographic and baseline characteristics were similar between subgroups; baseline MADRS total scores were distributed across both moderate and severe depressive symptom subgroups rather than clustering near severe MADRS total scores. For both subgroups, least squares mean treatment differences in change from baseline in HAMD-17 total scores favoured zuranolone at Day 15, with effects sustained to Day 45 (Table). Similar observations were observed for change from baseline in MADRS total scores (Table). There was no clear association between depressive symptom severity and the incidence of treatment-emergent adverse events within the placebo- and zuranolone-treated participants across both studies.

Conclusions

This post hoc analysis suggested a favourable treatment effect across subgroups with moderate or severe depressive symptoms based on baseline MADRS total scores. There was no clear difference in the incidence of treatment-emergent adverse events between subgroups. Overall, these post hoc analyses further characterise the efficacy and safety of zuranolone in participants with moderate or severe baseline depressive symptoms based on MADRS total scores in the SKYLARK and ROBIN studies.

Table.Least squares mean (LSM) treatment differences in change from baseline (CFB) in HAMD-17 and MADRS total scores by baseline MADRS total score severity.

	HAMD-17 total score LSM CFB	P-value	MADRS total score LSM CFB	P-value
SKYLARK, Day 15
Moderate	–4.4 (n=87)	0.005	–5.2 (n=87)	0.02
Severe	–4.0 (n=95)	0.03	–5.0 (n=94)	0.06
SKYLARK, Day 45
Moderate	–3.8 (n=80)	0.03	–4.5 (n=80)	0.05
Severe	–3.4 (n=88)	0.08	–5.0 (n=88)	0.08
ROBIN, Day 15
Moderate	–3.2 (n=60)	0.09	–3.1 (n=60)	0.20
Severe	–4.6 (n=87)	0.02	–5.7 (n=87)	0.04
ROBIN, Day 45
Moderate	–3.6 (n=58)	0.06	–4.6 (n=58)	0.06
Severe	–4.3 (n=84)	0.02	–7.1 (n=84)	0.01

References

1. Deligiannidis K.M., Meltzer-Brody S., Maximos B., et al., 2023. Zuranolone for the treatment of postpartum depression. Am J Psychiatry 180(9), 668-675.
2. Deligiannidis K.M., Meltzer-Brody S., Gunduz-Bruce H., et al., 2021. Effect of zuranolone vs placebo in postpartum depression: a randomized clinical trial. JAMA Psychiatry 78(9), 951-959.