Efficacy of tianeptine vs placebo in the long-term treatment (16.5 months)of unipolar major recurrent depression

To compare the efficacy and acceptability of tianeptine vs placebo in the l ong-term treatment of unipolar major recurrent depression, 268 hospitalized and ambulatory patients meeting DSM III-R criteria for major depression wi th a 21-item Hamilton depression rating scale (HDRS) score greater than or equal to 17 and at least one episode in the previous 5 years received tiane ptine in a 6-week multicenter open study. At D42, 185 responders (intention -to-treat population) were randomized for ethical reasons into two unbalanc ed groups to receive tianeptine 37.5 mg/day (n=111) or placebo (n=74) for 1 6.5 months; 173 of the 185 responders were defined as strict responders (pe r-protocol population) by an HDRS score which was both <15 and at least hal ved vs D1, combined with clinical confirmation by the investigator. The gro ups were similar at baseline except in the severity of the depressive episo de (greater in the tianeptine group: 33 vs 18%, p=0.018). Relapse (before 6 months) and recurrence (after 6 months) were defined by an HDRS score <gre ater than or equal to>15 and/or clinical global impression score greater th an or equal to4, and clinical confirmation by the investigator. Visits were at D63 and M3, M6, M9, M12, M15, and M18. Efficacy was measured by the num ber of relapses and recurrences and their time of onset (Kaplan-Meier survi val curve analysis). Between D42 and M18 (intention-to-treat population), r elapse and recurrence were less frequent on tianeptine us placebo (16 vs 36 %, p = 0.002). Comparison over time also showed a higher proportion of pati ents without relapse or recurrence on tianeptine (p <0.001); the intergroup difference increased with follow-up duration. Secondary analysis of relaps e in the intention-to-treat group showed a higher proportion on placebo (p= 0.002); secondary analysis of recurrence over time showed that the differen ce in the percentages of recurrence-free patients was nonsignificant in the intention-to-treat population (p = 0.067) but significant in the per-proto col population (p = 0.036) in favor of tianeptine. Acceptability did not di ffer between the groups. Treatment-induced adverse events were rare and mil d in both groups. These data support the use of tianeptine in the long-term treatment of unipolar major recurrent depression. Relapse and recurrence w ere decreased two- to threefold on tianeptine vs placebo with no difference in acceptability between the two groups. Copyright (C) 1997 Doin Editeurs.