To assess the Patient Health Questionnaire (PHQ-9) as a predictor of relapse of depressive symptoms in treatment-resistant depression (TRD).

Analysis included maintenance phase data from SUSTAIN-1 (NCT02493868), a randomized, double-blind, active-controlled study in TRD patients that evaluated efficacy of intranasal esketamine (ESK) + oral antidepressant (AD) vs AD + intranasal placebo in delaying relapse of depressive symptoms. A ≥50% reduction in initial symptom score and total score of ≤12 were considered as response and remission, respectively, using the Montgomery-Asberg Depression Rating Scale. PHQ-9 total score (range, 0–27), PHQ-2 total score (0–6), and individual items of the PHQ-9 (0–3) were examined as predictors of relapse. Data were collected every 2 weeks. Association between time-varying PHQ-9 and event of depression relapse was evaluated in Andersen-Gill Cox model.

Of 176 stable remitters, 63 had a relapse event (ESK+AD [n=24]; AD+placebo [n=39]). Of 121 stable responders, 50 had a relapse event (ESK+AD [n=16]; AD+placebo [n=34]). Among stable remitters, PHQ-9 total score (HR; 95% CI [1.12; 1.04–1.21]) and PHQ-2 total score (1.58; 1.25–1.99) were associated with relapse risk. PHQ-9 items #1 (loss of pleasure, 2.07; 1.38–3.09), #2 (feeling down, 2.18; 1.51–3.15), #4 (feeling tired, 1.54; 1.13–2.11), and #6 (negative self-view, 2.27; 1.41–3.66) were associated with relapse risk. PHQ-2 total scale yielded the smallest Akaike’s Information Criterion among stable remitters and responders.

PHQ-9, PHQ-2 total scores or individual items may be useful for predicting relapse of depressive symptoms among stable TRD patients.

This study was sponsored by Janssen Research and Development, LLC.