ObjectiveThis Phase II, placebo‐controlled, double‐blind study investigated the efficacy, safety, and tolerability of nivasorexant in the treatment of adults with moderate to severe binge‐eating disorder (BED).MethodsAdults meeting the DSM‐5 BED criteria were randomized 1:1 to placebo or nivasorexant (100 mg b.i.d.). The primary endpoint was the change from baseline to Week 12 in the number of binge eating (BE) days per week. Exploratory efficacy endpoints included cessation of BE in the last 4 weeks of treatment; and change from baseline to Week 12 in the number of BE episodes/week, the clinician global impression (CGI) of change, the Yale‐Brown Obsessive‐Compulsive Scale modified for BE, and the Hamilton rating scale for depression (HAMD‐17). Key safety outcomes included treatment‐emergent adverse events (TEAEs) and adverse events of special interest (i.e., somnolence and fatigue).ResultsSixty‐eight participants were randomized to each treatment arm. The change from baseline to Week 12 in the number of BE days/week was the same for placebo (least squares mean [LSM]: −2.93) and nivasorexant (LSM: −2.93), with no difference between the treatment groups (LSM difference = .000 [95% confidence interval (CI): −.69, .69], p = .9992). Furthermore, no differences between treatment groups were observed in the exploratory efficacy endpoints. Nivasorexant was well tolerated; the overall incidence of TEAEs was balanced between treatment groups, and the frequency of somnolence and fatigue in the nivasorexant group were similar to placebo.DiscussionIn this proof‐of‐concept study, 100 mg b.i.d. nivasorexant did not improve BE in adults with moderate to severe BED.Public SignificanceThe results of this Phase II study indicate that nivasorexant was well tolerated in adults with BED, but did not improve binge eating behavior over placebo. Further research is needed to improve our understanding of the role of the orexin‐1 receptor in BED.