To evaluate long-term efficacy and tolerability of the serotonin 5-HT1A receptor agonist, gepirone extended release (ER), a multicenter, randomized, placebo-controlled relapse prevention study was performed in patients with recurrent major depression (DSM-IV criteria). Patients 18 to 70 years, with a primary diagnosis of recurrent major depression (DSM-IV; 296.3) and a screening and baseline HAMD-17 total score >/=20 were eligible. After a 3- to 14-day (dependent on pretrial medication) single-blind placebo washout period, eligible patients entered an 8- or 12-week (depending on time to remission) open-label gepirone ER treatment period. They initially received a dose of 20 mg/d gepirone ER and were titrated to a dose of 40 to 80 mg/d. Patients who achieved remission (HAMD-17 total score </=8) were randomized to double-blind continuation of their gepirone ER treatment or placebo for 40 to 44 weeks. The primary end point was a comparison of the relapse rates between gepirone ER and placebo. Relapse was defined as a HAMD-17 total score >/=16 or discontinuation for lack of efficacy. A total of 420 patients were treated in the open-label phase. Of these, 303 (72.1%) completed the open-label phase and 250 (59.5%) fulfilled the criteria for remission and were randomized into the double-blind continuation phase (gepirone ER: n = 126; placebo: n = 124). The mean (+/-SD) final titrated dose of gepirone ER was 61.9 (+/-17.0) mg/d in the double-blind continuation phase. The relapse rate in the gepirone ER group was statistically significantly lower than that in the placebo group, 23.0% versus 34.7%, respectively (P = 0.024). During the open-label phase, adverse events that occurred in more than 5% of patients were nausea (15.7%), dizziness (13.1%), headache (12.9%), insomnia (6.2%), and vertigo (6.0%). During the continuation phase, the incidence of newly or re-emerging adverse events was similar with gepirone ER (43.7%) and placebo (42.7%). Adverse events different from those occurring during the open-label phase were not apparent. All adverse events occurred in less than 5% of patients with the exception of flu syndrome and headache. In conclusion, gepirone ER at a dose range of 40 to 80 mg/d is effective for relapse prevention in patients with recurrent major depression. It is well tolerated during long-term treatment for up to approximately one year.