European Neuropsychopharmacology (2014) 24, 553–563
www.elsevier.com/locate/euroneuro
A placebo-controlled study of three agomelatine dose regimens (10 mg, 25 mg, 25–50 mg) in patients with major depressive disorder$ Sidney H. Kennedya,n, Alla Avedisovab, Natalia Giménez-Montesinosc, Carole Belaïdic, Christian de Bodinatc, On behalf of agomelatine study group1 a
University of Toronto, University Health Network, Canada Serbsky State Scientific Center of Social and Forensic Psychiatry, Russia c Institut de Recherches Internationales Servier (IRIS), 50 rue Carnot, 92284 Suresnes Cedex, France b
Received 3 October 2013; received in revised form 9 January 2014; accepted 11 January 2014
KEYWORDS
Abstract
Agomelatine; Depression; Placebo; Dose regimen
A randomised placebo-controlled “dose relation study” was conducted in 549 patients who met the criteria for major depressive disorder, to evaluate the efficacy and safety of three doses regimens of agomelatine during 6 weeks: low fixed dosage (10 mg/day, n =133), fixed dosage (25 mg/day, n =138) and a flexible dosage with up-titration in case of insufficient improvement
Abbreviations: MDD, major depressive disorder ☆ Trial registration name: Efficacy and safety of 3 dose regimens of agomelatine (10, 25, 25–50 mg) versus placebo given once a day for 6 weeks in out-patients suffering from moderate to severe major depressive disorder. A 6-week randomised, double-blind, placebocontrolled, parallel groups study followed by a double-blind optional 18-week extension period. Trial registration number: EudraCT number: 2009-011238-84. n Correspondence to: FRCPC-University of Toronto EN8-222, 200 Elizabeth Street, Toronto General Hospital, Toronto, ON, Canada M5G 2C4. Tel.: +1 416 340 3888; fax: 1 416 340 4198. E-mail address: [email protected] (S.-n. Kennedy). 1 The agomelatine study CL3-20098-069 group of principal investigators: ARGENTINA: Dr. Ricardo Marcelo Corral; Dr. Mariano Alvarez Caches; Dr. Héctor Fabián Lamaison; Dr. Pedro Rafael Gargoloff; Dr. Christian Lupo ; Dr. Carlos Finkelsztein; Dr. Germán Mauricio Berardo; Dr. Daniel Mosca; Dr. Edith Mirta Serfaty; Dr. Enrique Kuper. BULGARIA: Pr. Vihra Milanova; Pr. Georgi Onchev; Pr. Radka Gaidarova; Pr. Georgi Popov; Dr. Loris Sayan; Dr. Svetlozar Georgiev; Pr. Valentin Akabaliev; Dr. Temenuzhka MatevA. FINLAND: Dr. Antti Ahokas; Dr. Jukka Penttinen; Dr. Marja Pirilä. RUSSIA: Pr. Alla Avedisova; Pr. Elena Grigorieva; Pr. Mikhail Ivanov; Pr. Lala Kasimova; Pr. Sergei Mosolov; Pr. Nikolay Neznanov; Pr. Anatoly Smulevich; Dr. Anna Vasilyeva; Pr. Oleg Shiryaev; Pr. Vladimir Tochilov. SLOVAKIA: Dr. Livia Vavrusova; Dr. Vladimir Garaj; Dr. Nada Kuriackova; Dr. Zuzana Janikova; Dr. Monika Biackova. UKRAINE: Pr. Oleg Chaban; Pr. Volodymyr Abramov; Pr. Olexandr Filts; Dr. Vladyslav Demchenko; Pr. Nataliya Maruta; Dr. Sergiy Mykhnyak; Pr. Oleksandr Napryeyenko; Pr. Valeriy Pidkorytov; Pr. Vitaliy Pishel. http://dx.doi.org/10.1016/j.euroneuro.2014.01.006 0924-977X & 2014 Elsevier B.V. and ECNP. All rights reserved.
554
S.H. Kennedy et al. at week 2 (25–50 mg/day, n =137). At last post-baseline assessment, there were significant and incremental placebo-agomelatine differences on mean HAM-D17 total scores in favour of each agomelatine dose regimen (2.4670.76 points, p=0.001 at 10 mg; 4.71 +0.75 points, po0.0001 at 25 mg and 4.9270.76 points, po0.0001 at 25–50 mg) with statistically significant differences between 25 mg and 25–50 mg dose regimens compared to the 10 mg dose. The response rate according to HAM-D17 was significantly higher in patients taking agomelatine than those taking placebo (difference of 16.1% at 10 mg p=0.005; 25.9% and 27.4% respectively at 25 mg and 25– 50 mg, po0.0001). The benefit of agomelatine was demonstrated in the subgroup of severely depressed patients in the 25 mg and 25–50 mg/day regimens. Consistent clinical response according to CGI variables and better social functioning were found in patients receiving agomelatine. All dose regimens of agomelatine were well tolerated and no unexpected adverse event was reported. This study provides evidence of a dose effect for agomelatine between 10 mg and the therapeutic dose regimen of agomelatine 25–50 mg: the efficacy of the higher dose regimens being more efficacious than the lowest (10 mg) daily dose. The data support a definitive statement regarding the utility of 25 mg as the threshold dose for initiating agomelatine in depressed patients. & 2014 Elsevier B.V. and ECNP. All rights reserved.
1.
Introduction
When selecting a first-line antidepressant agent, considerations include the potential to achieve a greater response with increasing dosages, thereby allowing greater flexibility in dosing. Indeed, increasing the dose of antidepressants seems to be the most straightforward strategy when depressed patients have an insufficient response after adequate time (generally at least 8 weeks) at the initial therapeutic dose (Fredman et al., 2000). However, for most available antidepressants, there are few data from randomised controlled trials to support or refute this approach (Berney, 2005). In the case of SSRI and SNRI antidepressants, it has been difficult to establish a response relationship (Beasley Jr. et al., 1990; Bech et al., 2002; Burke et al., 2002; Dunner and Dunbar, 1992; Fabre et al., 1995; Fabre and Putman III, 1987; Khan et al., 1998; Mallinckrodt et al., 2006; Mendels et al., 1993; Montgomery et al., 1993, 1994; Rudolph et al., 1998; Ruhe et al., 2009; Wernicke et al., 1988). The threshold dose of agomelatine, an antidepressant acting as a MT1 and MT2 receptor agonist and a 5-HT2C receptor antagonist (de Bodinat et al., 2010), is 25 mg/day. This active dose was established in an 8-week double-blind placebo-controlled dose finding study evaluating three dose regimens (1, 5 and 25 mg once a day) (Loo et al., 2002) and several successful placebo controlled or head to head trials have subsequently verified the therapeutic effectiveness of both fixed 25 mg and one step titration 25–50 mg dose regimens (de Bodinat et al., 2010; Hickie and Rogers, 2011; Kennedy and Rizvi, 2010). Following the European Medical Agency requirement at the time of the centralised European Marketing Authorisation, a post-commitment dose relation study was scheduled to ascertain the threshold dose for the therapeutic efficacy of agomelatine in depressed patients and provide evidence for a dose effect of agomelatine. The purpose of this study was to evaluate the efficacy and safety of low dose agomelatine compared to standard doses
of agomelatine in a randomised, placebo controlled trial involving patients with major depressive disorder (MDD). A 10 mg low dose was selected as an intermediate dose between 5 mg (previously assessed in Loo et al. (2002)) and 25 mg and was compared to currently recommended dosages of 25 mg and 25 mg uptitrated to 50 mg for patients with an insufficient improvement after 2 weeks.
2.
Procedures
The study was initiated only after the written Ethics Committee's approval, in accordance with the local regulations in each of the countries, and was conducted in accordance with the principles of Good Clinical Practice and the Declaration of Helsinki. Subjects signed informed consent prior to participating in the trial.
2.1.
Patients
Male or female outpatients aged between 18 and 65 years (inclusive) with a primary diagnosis of MDD of moderate to severe intensity according to DSM-IV-TR (American Psychiatric Association, 2000) and a current major depressive episode (MDE) confirmed by the Mini International Neuropsychiatric Interview (Sheehan et al., 1998) were recruited in 45 centres in 6 countries: Argentina (102 patients), Bulgaria (95 patients), Finland (51 patients), Russia (156 patients), Slovakia (61 patients), and Ukraine (84 patients) between October 2009 and November 2011. Patients had to have a Hamilton Depression Rating Scale 17 items (HAM-D17) total score Z 22, a score of 44 on the item 1 (severity of illness) of the Clinical Global Impression (CGI) scale (Guy, 1976), a Hospital and Depression Anxiety (HAD) (Zigmond and Snaith, 1983) depression subscore of Z 11. Subjects with a decrease of greater than 20% on the HAM-D17 between selection and inclusion were excluded. All patients were required to be physically healthy or to have stabilised somatic illnesses.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder Patients were excluded if they had an ongoing episode of MDD of less than 4 week-duration or with seasonal pattern, psychotic or catatonic features, post-partum onset or onset after stroke within 3 months; if they had resistant depression or no previous response to agomelatine treatment; if they were at risk of suicide; if they had been treated with electroconvulsive therapy (ECT) within the previous 3 months; if they had insight-oriented and structured psychotherapy within the previous 3 months, or lighttherapy within the previous 2 weeks. All types of depression other than MDD according to DSM-IV-TR criteria were excluded as well as current panic disorder, obsessive compulsive disorder, post-traumatic stress disorder, acute stress disorder, acute or chronic psychosis, antisocial, borderline or histrionic personality disorders or any other severe personality disorder (paranoid, schizoid, narcissistic, avoidant, dependent) or alcohol or drug abuse or dependence within the past 12 months. Apart from agomelatine, patients did not receive any other psychotropic treatment during the study, or any treatment prone to interfere with the central nervous system.
2.2.
Treatment allocation
Using an Interactive Response System and according to a balanced (non-adaptive) randomisation with stratification for each centre, patients were assigned to receive agomelatine 10 mg, agomelatine 25 mg, agomelatine 25 mg increased up to 50 mg (one step titration dose regimen), or placebo. For the one step titration regimen, insufficient improvement after 2 weeks led to an increase in the dosage of agomelatine to 50 mg daily, according to a predefined dose adjustment algorithm. Both investigators and subjects remained blind to the up-titration.
2.3.
Measures
The primary outcome measure was the HAM-D17 total score, rated at the selection visit and at weeks 0, 2, 4, and 6. The primary outcome variable was the last post-baseline HAMD17 total score over 6 weeks. Secondary outcome measures were the HAM-D17 response rate (defined as a Z 50% decrease in HAM-D17 total score from baseline), CGI-S score at each visit from selection to week 6 and CGI-I at each visit from week 2 to week 6, CGI-I response rate (defined as CGI-I score = 1 or 2), and Sheehan Disability Scale (SDS) scores at selection and week 6. Safety measures included adverse events reported at each visit, 12-lead ECGs and laboratory tests (biochemistry, haematology) at selection and week 6. A physical examination (blood pressure, heart rate, weight and body mass index) was performed at selection, weeks 0 and 6. If the patient withdrew prematurely, the above-mentioned efficacy and safety measurements were repeated at the time of withdrawal.
2.4.
Statistical analysis
The efficacy analyses were performed on the Full Analysis Set (FAS, randomised patients who took at least one dose of medication, with a value at baseline and at least one post-
555
baseline value for the primary efficacy criterion). The primary analysis examined agomelatine-placebo differences on the last post-baseline value of the HAM-D17 total score over the 6-week period, by using a two-way analysis of covariance (ANCOVA) model on factor treatment (including the 4 treatment groups) with centre (random effect), and baseline HAM-D17 total score as covariates and without interaction. The Hochberg procedure was used to deal with multiplicity issues, providing Hochberg-adjusted p-values. A sensitivity analysis to the method of handling missing values was performed in the FAS. Treatment groups were compared to placebo on the value at week 6, using a mixedeffects model for repeated measures (MMRM), including terms for the fixed effects of treatment, baseline HAM-D17 total score, visit and an interaction term for treatment and visit, and for the random effect of centre. The analysis fitted an unstructured covariance matrix. In the same way as for the main analysis, the Hochberg procedure was also applied. Additional analyses were conducted in the FAS using a χ2 test to assess agomelatine-placebo differences in response to treatment taking into account the last postbaseline value over 6 weeks. The same analyses were also repeated in the planned subset of patients with both HAM-D17 total score Z25 and CGI-S scoreZ 5 at baseline (severely ill patients). Treatment groups were compared to placebo over the 6week period in the FAS for (1) the last post-baseline value of CGI-S score and last value of CGI-I score by using a two-sided Student's t-test for independent samples and a Mann– Whitney test; (2) the response to treatment according to the CGI-I (global improvement score = 1 or 2), using a χ2 test on the last value and (3) SDS work/daily activities, social life and family life scores, number of days lost and number of unproductive days taking into account the mean change from baseline to last post-baseline value using a two-sided Student's t-test (post-hoc analysis). The difference on the primary outcome measures between agomelatine 10 mg versus agomelatine 25 mg and agomelatine 25–50 mg was estimated in the FAS using a two-way ANCOVA model, as for the main analysis but including only the 3 agomelatine treatment groups (post-hoc analysis). For every safety measurement, descriptive statistics were provided by treatment group in the safety set (all included patients who took at least one dose of study medication). Statistical analysis was performed using the SASs software, version 9.1 or 9.2 (SAS Inc.; Cary, North Carolina). The type I error was set at 5% (two-sided tests).
3. 3.1.
Results Patients
Five hundred and forty nine patients were randomised to receive agomelatine 10 mg (133 patients), agomelatine 25 mg (138 patients), a one step titration dosage of agomelatine 25–50 mg (137 patients) or placebo (141 patients). A total of 505 patients completed the 6-week treatment period (91.9%). Reasons for premature withdrawal were mainly lack of efficacy or non-medical reason (Figure 1). There were no clinically relevant differences
556
S.H. Kennedy et al.
Screened N = 683
Not Randomized Not fulfilled study entry criteria Adverse events Withdrew consent
N = 134 n=116 n = 10 n=8
Randomized N = 549
Agomelatine 10 mg N = 133
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
n=123 n=10 n=1 n=2 n=5 n=2
Agomelatine 25 mg fixed N = 138
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
Figure 1
n=130 n=8 n=1 n=1 n=3 n=3
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
n=126 n=11 n=1 n=1 n=5 n=4
Placebo N = 141
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
n=126 n=15 n=0 n=0 n=9 n=6
Disposition of patients.
between the groups for demographic and clinical characteristics (Table 1). In the group of patients who were eligible for up-titration, 49 out of 134 patients (36.6%) had a dose increase of agomelatine from 25 mg to 50 mg after 2 weeks.
3.2.
Agomelatine 25 or 25-50 mg N = 137
Efficacy
3.2.1. Whole study population The mean HAM-D17 total score decreased from baseline to week 6 in all groups (Figure 2). Agomelatine was associated with a clinically relevant decrease in symptoms at the last post-baseline value (Table 2). The placebo-agomelatine differences ranged from 2.4670.76 points (p= 0.001) in the 10 mg group, to 4.7170.75 (po0.0001) in the 25 mg group and 4.9270.76 points (po0.0001) in the group of patients with potential one step titration 25–50 mg dosage regimen. Results from the MMRM sensitivity analysis were consistent with the main analysis. The mean HAM-D17 total score at the last post-baseline assessment was not statistically different between the agomelatine 25 mg and 25–50 mg groups (E(SE)= 0.21 (0.67), 95% CI= [ 1.12; 1.54]). The mean last postbaseline HAM-D17 total score in the agomelatine 10 mg group was statistically significantly smaller than in both agomelatine 25 mg (E(SE) = 2.24 (0.72), 95% CI= [0.82; 3.65]) and agomelatine 25–50 mg groups (E(SE)= 2.44 (0.72), 95% CI= [1.02; 3.86], post-hoc analysis). Regarding secondary efficacy outcomes, the response rate according to HAM-D17 was significantly higher in patients taking agomelatine than those taking placebo, with a clinically relevant difference from placebo of 16.1% on agomelatine 10 mg (p= 0.005), 25.9% on agomelatine 25 mg (po0.0001), and 27.4% on agomelatine 25–50 mg (po0.0001) (Table 2). The differences of the mean CGI-S
and CGI-I scores between the 3 agomelatine dose regimens and placebo were statistically significant at the last (postbaseline) value with the Student's t-test (except for a trend to statistical significance for agomelatine 10 mg (p= 0.076) for the mean CGI-S score) and with the Mann–Whitney test (Table 2). The percentage of responders according to CGI-I was significantly higher in the three agomelatine groups (from 52.3% to 71.7%) than in the placebo group (36.9%) (Table 2). Over the first 6 weeks of treatment, mean decreases in the three SDS sub-scales; the mean decrease in number of lost work days, and number of underproductive days at the last post-baseline assessment were significantly greater in both fixed 25 mg and one step titration 25–50 mg agomelatine groups, than in the placebo group. In contrast, in the agomelatine 10 mg group, the effects on “Work and activity” score and the number of underproductive days, were not significantly different from those obtained in the placebo group (Table 3).
3.2.2. Severely ill patients The mean HAM-D17 total score decreased from baseline to week 6 in all groups (Figure 3). In the subset of severely depressed patients with baseline HAM-D17 total score Z25 and CGI-S Z 5, at the last post-baseline value, the placeboagomelatine differences on the HAM-D17 total score did not achieve statistical significance at the 10 mg regimen dosage (1.5771.16, p =0.176) but was highly significant at fixed 25 mg (5.3871.16, po0.0001) and at the one step titration 25–50 mg (5.5371.20, po0.0001) dosages (Table 4). Results from the MMRM sensitivity analysis were consistent with the main analysis. The response rate according to HAM-D17 was significantly higher in patients on agomelatine 25 mg fixed and one step titration 25–50 mg dosage than on placebo,
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder
Table 1
557
Demographics and baseline characteristics (randomised set).
Characteristics
Agomelatine
Age, mean7SD [range] Gender, female n (%) DSM-IV MDE single, n (%) MDE recurrent, n (%) Current MDE duration before inclusion (months), median HAM-D-17 total score, mean7SD CGI-S score, mean7SD
10 mg (N =133)
25 mg fixed (N =138)
25–50 mg (N =137)
Placebo (N =141)
45.4712.6 [18;65] 105 (79.0)
46.5711.9 [20;65] 90 (65.2)
43.1712.6 [19;65] 98 (71.5)
45.0713.0 [18;65] 108 (76.6)
33 (24.8) 100 (75.2) 2.50
30 (21.7) 108 (78.3) 2.50
34 (24.8) 103 (75.2) 2.50
35 (24.8) 106 (75.2) 2.50
27.272.7 4.770.6
26.772.9 4.770.6
26.772.9 4.670.6
26.672.6 4.670.7
Abbreviations: MDE=major depressive episode; DSM-IV=diagnostic and statistical manual of mental disorders (fourth edition, Text Revision); HAM-D-17=Hamilton depression rating scale 17 items; CGI-S=Clinical Global Impression Severity of illness score.
30
RANDOMISATION TREATMENT
29
Agomelatine 10 mg Agomelatine 25 or 25-50 mg
Agomelatine 25 mg fixed Placebo
28 27 26 25 24
MEAN VALUE
23 22 21 20 19 18 17 16 15 14 13 12 11 10 W000
W002
W004
W006
LAST POST-BASELINE (ON W000-W006)
VISIT
Figure 2
Course of the HAM-D score over the time of 6 weeks for all four treatment-groups in the whole study population.
with a clinically relevant placebo-agomelatine difference of 25.0% and 25.8%, respectively (p= 0.002 in each group respectively).
3.3.
Tolerability
During the 6-week study period, in the safety set, the percentage of patients with at least one emergent adverse event was higher in the 3 agomelatine groups (10 mg [24.2%], 25 mg [27.5%] and one step titration 25–50 mg [35.3%]) than in the placebo group (18.4%). Headache and
nausea were the most frequent emergent adverse events (EAEs) reported by patients in agomelatine groups (Table 5). The majority of adverse events were rated as mild or moderate. In all, 3 patients had severe emergent adverse events (1 patient each in the agomelatine 10 mg, 25 mg fixed and placebo groups). The percentage of patients who experienced at least one emergent adverse event leading to study treatment discontinuation was 0.8% (1 patient) in the agomelatine 10 mg group, 0.7% (1 patient) in the agomelatine 25 mg fixed group, 1.5% (2 patients) in the agomelatine one step titration 25–50 mg group, and 0.7% (1 patient) in the placebo group. Serious adverse events reported in one
558
Table 2
S.H. Kennedy et al.
Primary and key secondary efficacy outcomes over 6 weeks (Full Analysis Set). Agomelatine 10 mg (N =132)
25 mg fixed (N =138)
25–50 mg (N =136)
Placebo (N =141)
HAM-D-17 total score (a) n Baseline, mean7SD Last post-baseline value, mean7SD Difference from placebo (SE)a 95% CI p-Value b
132 27.372.6 16.577.0 2.46 (0.76) [0.96;3.96] 0.001 (III)
138 26.772.9 14.076.6 4.71 (0.75) [3.23;6.19] o0.0001 (II)
136 26.772.9 13.976.3 4.92 (0.76) [3.43;6.40] o0.0001 (I)
141 26.672.6 18.777.3
MMRM (b) n at week 6 Value at week 6, mean7SD Difference from placebo (SE)a 95% CI p-Valueb
126 16.176.8 2.83 (0.77) [1.32;4.34] o0.001 (III)
131 13.776.4 4.81 (0.76) [3.32;6.30] o0.0001 (II)
130 13.375.7 5.08 (0.76) [3.58;6.57] o0.0001 (I)
129 18.077.0
HAM-D-17 response rate (c) n Last post-baseline value, n (%) Difference from placebo (SE)c 95% CI p-Valued
132 54 (40.9) 16.1 (5.6) [ 27.1; 5.1] 0.005
138 70 (50.7) 25.9 (5.6) [ 36.9; 14.9] o0.0001
136 71 (52.2) 27.4 (5.6) [ 38.4; 16.4] o0.0001
141 35 (24.8)
CGI-S score (d) n Last post-baseline value, mean7SD Difference from placebo (SE)e 95% CI p-Valuef p-Valueg
132 3.571.1 0.25 (0.14) [ 0.03;0.52] 0.076 0.044
138 3.171.1 0.67 (0.13) [0.41;0.92] o0.0001 o0.0001
136 3.171.0 0.65 (0.13) [0.39;0.91] o0.0001 o0.0001
141 3.771.1
CGI-I score (d) n Last value, mean7SD Difference from placebo (SE) 95% CI p-Valuef p-Valueg
132 2.671.0 0.33 (0.13) [0.07;0.59] 0.013 0.009
138 2.271.0 0.76 (0.13) [0.51;1.01] o0.0001 o0.0001
136 2.270.9 0.77 (0.12) [0.52;1.01] o0.0001 o0.0001
141 2.971.1
CGI-I response rate (c) n Last value, n (%) Difference from placebo (SE) 95% CI p-Value d
132 69 (52.3) 15.4 (6.0) [ 27.1; 3.7] 0.011
138 99 (71.7) 34.9 (5.6) [ 45.8; 23.9] o0.0001
136 97 (71.3) 34.4 (5.6) [ 45.5; 23.4] o0.0001
141 52 (36.9)
(a) Analysis of covariance model on factor treatment with baseline HAM-D-17 total score and centre (random effect) as covariates; (b) mixed-effects model with repeated measures including terms for effects of treatment, baseline HAM-D-17 total score, centre (random effect), visit and an interaction term treatment visit; (c) χ2 test; and (d) two-sided Student's t-test for independent samples/ Mann–Whitney test. CGI-S=Clinical Global Impression Severity of illness score; CGI-I =clinical global impression global improvement score; HAM-D17=Hamilton depression rating scale 17 items a Difference between adjusted treatment group means (placebo minus agomelatine). b Two-sided Hochberg-adjusted p-value with (I)/(II)/(III) corresponding to the ordered unadjusted p-values from the most significant to the least significant one. c Difference between percentages (Placebo minus Agomelatine). d Two-sided p-value. e Difference between treatment group means (Placebo minus Agomelatine). f Student's t-test: two-sided p-value. g Mann–Whitney test: two-sided p-value.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder
Table 3
559
Mean SDS change from baseline to last post-baseline assessment over 6 weeks (Full Analysis Set). Agomelatine 10 mg (N =132)
25 mg fixed (N =138)
Work and activitya n Mean7SD Difference from placebo (SE)b 95% CI p-Value c
98 2.172.2 0.48 (0.30) [ 0.11;1.08] 0.112
102 2.972.6 1.32 (0.33) [0.68;1.97] o0.0001
103 2.972.2 1.36 (0.29) [0.79;1.94] o0.0001
102 1.672.0
Social life n Mean7SD Difference from placebo (SE) 95% CI p-Valuec
127 2.572.4 0.78 (0.28) [0.23;1.33] 0.005
127 3.072.6 1.25 (0.29) [0.67;1.83] o0.0001
128 3.172.5 1.34 (0.29) [0.78;1.90] o0.0001
136 1.772.2
Family life and home responsibilities n 127 Mean7SD 2.572.4 Difference from placebo (SE) 0.73 (0.29) 95% CI [0.16;1.30] p-Valuec 0.012
127 3.072.6 1.28 (0.30) [0.68;1.88] o0.0001
128 3.172.4 1.35 (0.29) [0.78;1.93] o0.0001
135 1.772.3
Number of days lost n Mean7SD Difference from placebo (SE) 95% CI p-Valuec
120 1.472.3 0.54 (0.27) [0.01;1.07] 0.046
120 1.972.6 0.98 (0.28) [0.42;1.54] o0.001
127 0.971.9
121 2.472.5 1.06 (0.32) [0.42; 1.69] 0.001
121 2.872.6 1.39 (0.32) [0.75; 2.03] o0.0001
126 1.472.5
123 1.472.2 0.52 (0.26) [0.02;1.03] 0.044
Number of underproductive days n 126 Mean7SD 1.872.8 Difference from placebo (SE) 0.40 (0.33) 95% CI [ 0.26; 1.05] 0.236 p-Valuec
25–50 mg (N =136)
Placebo (N =141)
a
Only values of patients having worked/studied during the past week. Difference between treatment group means (Placebo minus Agomelatine). c Two-sided Student's t-test for independent samples. b
patient (0.7%) receiving agomelatine 25 mg (depression, insomnia, anxiety, restlessness and decreased appetite) and one patient (0.7%) in the placebo group (haemorrhagic enterocolitis) were considered as unrelated to treatment by the investigator. No death occurred during the study. EAEs related to study treatment were reported more often by patients taking agomelatine (10.6%, 13.8%, and 15.4% in the agomelatine 10 mg, 25 mg fixed, and one step titration 25–50 mg groups, respectively) than by those taking placebo (6.4%). There were no clinically relevant between-group differences or changes from baseline to the last value during treatment for weight and vital signs, blood pressures and heart rate, biochemical and haematological parameters. No clinically relevant ECG abnormalities were recorded. During the 6-week treatment period of the study, no
emergent potentially clinically significant abnormal transaminase levels (43 ULN) were observed (Table 6).
4.
Discussion
This placebo-controlled study demonstrates the clinical efficacy and safety of three dose regimens of agomelatine given once a day for 6 weeks in out-patients suffering from moderate to severe major depressive disorder. Each dose regimen arm separated significantly from placebo with relevant differences of 2.46, 4.71 and 4.92 points on the final HAM-D17 total score respectively for the 10 mg, 25 mg, and 25–50 mg agomelatine groups. The antidepressant efficacy is supported by consistent differences in rates of response compared to placebo; 16% agomelatine-placebo
560
S.H. Kennedy et al. 30 RANDOMISATION TREATMENT
29
Agomelatine 10 mg Agomelatine 25 or 25-50 mg
Agomelatine 25 mg fixed Placebo
28 27 26 25 24
MEAN VALUE
23 22 21 20 19 18 17 16 15 14 13 12 11 10 W000
W002
W004
W006
LAST POST-BASELINE (ON W000-W006)
VISIT
Figure 3 Course of the HAM-D score over the time of 6 weeks for all four treatment-groups in the subset of severely ill patients. Table 4 Efficacy outcomes over 6 weeks in the more severely depressed patients at inclusion (Sub-FAS with baseline HAM-D17 total score Z25 and CGI-S Z5). Agomelatine 10 mg (N =132)
25 mg fixed (N =138)
25 50 mg (N =136)
Placebo (N =141)
HAM-D-17 total score (a) n Baseline, mean7SD Last post-baseline value, mean7SD Difference from placebo (SE)a 95% CI p-Valueb
73 28.572.2 18.377.0 1.57 (1.16) [ 0.71; 3.85] 0.176 (III)
70 28.272.9 14.277.0 5.38 (1.16) [3.10; 7.67] o0.0001 (I)
63 28.772.8 14.376.6 5.53 (1.20) [3.18; 7.89] o0.0001 (II)
68 28.272.6 19.677.7
MMRM (b) n at week 6 Value at week 6, mean7SD Difference from placebo (SE)a 95% CI p-Valueb
71 18.176.9 1.72 (1.14) [ 0.53; 3.97] 0.133 (III)
68 14.277.0 5.27 (1.15) [3.01;7.53] o0.0001 (II)
59 13.475.8 5.74 (1.19) [3.40;8.07] o0.0001 (I)
61 18.777.5
HAM-D-17 response rate (c) n Last post-baseline value, n (%) Difference from placebo (SE)c 95% CI p-Valued
73 26 (35.6) 10.6 (7.7) [-25.7;4.4] 0.171
70 35 (50.0) 25.0 (8.0) [ 40.6; 9.4] 0.002
63 32 (50.8) 25.8 (8.2) [ 41.9; 9.7] 0.002
68 17 (25.0)
(a) Analysis of covariance model on factor treatment with baseline HAM-D-17 total score and centre (random effect) as covariates; (b) mixed-effects model with repeated measures including terms for effects of treatment, baseline HAM-D-17 total score, centre (random effect), visit and an interaction term treatment visit; (c) χ2 test. a Difference between adjusted treatment group means (Placebo minus Agomelatine). b Two-sided Hochberg-adjusted p-value with (I)/(II)/(III) corresponding to the ordered unadjusted p-values from the most significant to the least significant one. c Difference between percentages (Placebo minus Agomelatine). d Two-sided p-value.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder
561
Table 5 Emergent adverse events during the double-blind 6-week treatment period in at least 2% of any agomelatinetreated patients (Safety Set). Agomelatine
Headache Nausea Nasopharyngitis Diarrhoea Dizziness Somnolence Constipation Dry mouth
10 mg (N= 132)
25 mg fixed (N =138)
25–50 mg (N =136)
Placebo (N= 141)
11 (8.3) 3 (2.3) – 1 (0.8) – 3 (2.3) – 1 (0.8)
6 7 – 2 2 1 3 3
10 (7.4) 5 (3.7) 4 (2.9) 4 (2.9) 4 (2.9) 3 (2.2) 1 (0.7) 1 (0.7)
7 2 4 – – 2 1 –
(4.3) (5.1) (1.4) (1.4) (0.7) (2.2) (2.2)
(5.0) (1.4) (2.8)
(1.4) (0.7)
Number of patients (%).
Table 6 ALT/AST parameters during the double-blind 6-week treatment period. Mean changes from baseline to last value (Safety Set). Agomelatine
ASAT (IU/L) ALAT (IU/L)
n Mean7SD n Mean7SD
10 mg (N =132)
25 mg fixed (N =138)
25–50 mg (N =136)
Placebo (N =141)
125 0.676.2 125 0.379.7
129 0.177.4 129 1.3711.3
128 0.475.8 128 0.1710.0
134 0.376.5 134 0.3710.9
difference on agomelatine 10 mg; 26% and 27% on agomelatine 25 mg and 25–50 mg respectively and similar difference on CGI variables. This study also confirms a dose effect of agomelatine, with both superiority of the 10 mg dose over placebo and superiority of agomelatine 25 mg and 25–50 mg over both agomelatine 10 mg and placebo. Efficacy of the two higher dose regimens (25 mg and 25–50 mg) of agomelatine compared to placebo is further substantiated in the predefined ‘severe depression’ subpopulation, who displayed a 5 point agomelatine-placebo difference at endpoint according to HAM-D17 total score and a 25% difference from placebo in clinical response. Thus, those patients likely to be most impaired by their depression and most likely to receive antidepressant treatment in routine practice need a minimum dosage of 25 mg daily. There is a bilateral relationship between depressive symptoms, reduced social capabilities and functional impairment. Based on SDS findings, agomelatine helps to improve symptom-related functional impairment in patients over 6 weeks, and those improvements were noted on all domains for the fixed dose 25 mg, and the one step titration 25–50 mg dose regimen. This translates into better social functioning and quality of life for patients, as has been previously reported at both fixed 25 mg and one step titration 25– 50 mg dosages (Goodwin, 2009; Heun et al., in press). In contrast, the absence of significant effects on “Work and activity” score and ‘number of underproductive days’ in the agomelatine 10 mg group provides further evidence that the
lowest tested dose displays lower efficacy when compared to agomelatine 25 mg with or without up-titration to 50 mg. Finally, the three dose regimens of agomelatine were well tolerated during the 6-week treatment period. Compared to the placebo group, higher incidences of emergent adverse events were reported in the 3 agomelatine groups, and this is compatible with the observation that side-effects commonly increase with increased dosing. However, no unexpected adverse event occurred, supporting previous reports on the side effect profile of agomelatine (de Bodinat et al., 2010; Hickie and Rogers, 2011; Kennedy and Rizvi, 2010), and there were no specific concerns about liver enzymes on the short-term in this study. These findings are noteworthy as they represent the first demonstration of a dose response effect of an antidepressant medication, with statistically significant differences between doses. To date, when reviewing the literature, only equivocal findings were obtained regarding a dose–response relationship for all currently available antidepressants, notably among the SSRIs, fluoxetine (Fabre and Putman III, 1987; Wernicke et al., 1988; Beasley Jr. et al., 1990), citalopram (Bech et al., 2002; Montgomery et al., 1994; Montgomery et al., 1993), escitalopram (Burke et al., 2002), paroxetine (Dunner and Dunbar, 1992; Ruhe et al., 2009) and sertraline (Fabre et al., 1995). Neither was a dose-relationship demonstrated for venlafaxine (Khan et al., 1998; Mendels et al., 1993; Rudolph et al., 1998) or duloxetine (Mallinckrodt et al., 2006). The dose effect for agomelatine between 10 mg and therapeutic dose regimen of agomelatine 25–50 mg complements
562 and extends the original dose finding study by Loo et al. (2002), in which 25 mg proved the most efficacious compared to 1 mg and 5 mg and represents the threshold dose for initiating agomelatine in a secure way for the patients. The clinically relevant findings obtained with the one step titration 25–50 mg dose also confirm the utility of 50 mg for some patients. In routine clinical practice, when no early treatment improvement is observed, the clinician classically increases the dose of treatment before considering a change in therapeutic management.
Role of the funding source This study was sponsored by Servier (Suresnes, France). The sponsor of the study participated in study design, data collection, data analysis, data interpretation, and writing of the report. The authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Contributors The sponsor of the study participated in study design, data collection, data analysis, data interpretation, and writing of the report. The authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis
Conflict of interest Sidney Kennedy has acted as advisor/consultant or speaker for AA Pharma, Abbott Laboratories, AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Elsevier, Forest Laboratories, Glia Communications, Indigene Lifesystems Pvt. Ltd., Lundbeck, Lundbeck Institute, Pfizer, Servier, Spimaco, Symbiote Medical Communications Inc. and Spirant Communications. He has received grant support from Bristol-Myers Squibb, Brain Cells Inc., Canadian Institutes of Health Research, Clera Inc., Eli Lilly, Lundbeck, Ontario Brain Institute, Pfizer, Servier and St. Jude Medical Inc. Alla Avedisova has received grant/research support from GlaxoSmithKline, Lundbeck, Sanofi and Servier; has served as a consultant or on advisory boards for Janssen and Servier Natalia Giménez-Montesinos, Carole Belaïdi and Christian de Bodinat are employees at Servier.
Ackowledgement None.
References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. American Psychiatric Association, Washington, DC. Beasley Jr., C.M., Bosomworth, J.C., Wernicke, J.F., 1990. Fluoxetine: relationships among dose, response, adverse events, and plasma concentrations in the treatment of depression. Psychopharmacol. Bull. 26, 18–24. Bech, P., Tanghoj, P., Andersen, H.F., Overo, K., 2002. Citalopram dose–response revisited using an alternative psychometric approach to evaluate clinical effects of four fixed citalopram
S.H. Kennedy et al. doses compared to placebo in patients with major depression. Psychopharmacology (Berl.) 163, 20–25. Berney, P., 2005. Dose–response relationship of recent antidepressants in the short-term treatment of depression. Dialog. Clin. Neurosci. 7, 249–262. Burke, W.J., Gergel, I., Bose, A., 2002. Fixed-dose trial of the single isomer SSRI escitalopram in depressed outpatients. J. Clin. Psychiatry 63, 331–336. de Bodinat, C., Guardiola-Lemaitre, B., Mocaer, E., Renard, P., Munoz, C., Millan, M.J., 2010. Agomelatine, the first melatonergic antidepressant: discovery, characterization and development. Nat. Rev. Drug Discov. 9, 628–642. Dunner, D.L., Dunbar, G.C., 1992. Optimal dose regimen for paroxetine. J. Clin. Psychiatry 53 (Suppl.), S21–S26. Fabre, L.F., Abuzzahab, F.S., Amin, M., Claghorn, J.L., Mendels, J., Petrie, W.M., Dube, S., Small, J.G., 1995. Sertraline safety and efficacy in major depression: a double-blind fixed-dose comparison with placebo. Biol. Psychiatry 38, 592–602. Fabre, L.F., Putman III, H.P., 1987. A fixed-dose clinical trial of fluoxetine in outpatients with major depression. J. Clin. Psychiatry 48, 406–408. Fredman, S.J., Fava, M., Kienke, A.S., White, C.N., Nierenberg, A.A., Rosenbaum, J.F., 2000. Partial response, nonresponse, and relapse with selective serotonin reuptake inhibitors in major depression: a survey of current "next-step" practices. J. Clin. Psychiatry 61 (6), 403–408. Goodwin, G.M., 2009. Clinical studies on the efficacy of agomelatine on depressive symptoms. CNS Drugs 23 (Suppl 2), S35–S39. Guy, W., 1976. ECDEU Assessment Manual for Psychopharmacology. US Department of Health, Education and Welfare Publication ADM 76-338. National Institute of Mental Health, Rockville, MD217–222. Heun, R., Ahokas, A., Boyer, P., Giménez-Montesinos, N., PontesSoares, F., Olivier, V., 2014. The efficacy of agomelatine in elderly patients with major recurrent depressive disorder: a placebo controlled study. J. Clin. Psychiatry. Hickie, I.B., Rogers, N.L., 2011. Novel melatonin-based therapies: potential advances in the treatment of major depression. Lancet 378, 621–631. Kennedy, S.H., Rizvi, S.J., 2010. Agomelatine in the treatment of major depressive disorder: potential for clinical effectiveness. CNS Drugs 24, 479–499. Khan, A., Upton, G.V., Rudolph, R.L., Entsuah, R., Leventer, S.M., 1998. The use of venlafaxine in the treatment of major depression and major depression associated with anxiety: a dose–response study. Venlafaxine investigator study group. J. Clin. Psychopharmacol. 18, 19–25. Loo, H., Hale, A., D'haenen, H, 2002. Determination of the dose of agomelatine, a melatoninergic agonist and selective 5-ht2c antagonist, in the treatment of major depressive disorder: a placebo-controlled dose range study. Int. Clin. Psychopharmacol. 17, 239–247. Mallinckrodt, C.H., Prakash, A., Andorn, A.C., Watkin, J.G., Wohlreich, M.M., 2006. Duloxetine for the treatment of major depressive disorder: a closer look at efficacy and safety data across the approved dose range. J. Psychiatr. Res. 40, 337–348. Mendels, J., Johnston, R., Mattes, J., Riesenberg, R., 1993. Efficacy and safety of b.i.d. doses of venlafaxine in a dose–response study. Psychopharmacol. Bull. 29, 169–174. Montgomery, S.A., Pedersen, V., Tanghoj, P., Rasmussen, C., Rioux, P., 1994. The optimal dosing regimen for citalopram—a metaanalysis of nine placebo-controlled studies. Int. Clin. Psychopharmacol. 9 (Suppl. 1), S35–S40. Montgomery, S.A., Rasmussen, J.G., Tanghoj, P., 1993. A 24-week study of 20 mg citalopram, 40 mg citalopram, and placebo in the prevention of relapse of major depression. Int. Clin. Psychopharmacol. 8, 181–188.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder Rudolph, R.L., Fabre, L.F., Feighner, J.P., Rickels, K., Entsuah, R., Derivan, A.T., 1998. A randomized, placebo-controlled, dose– response trial of venlafaxine hydrochloride in the treatment of major depression. J. Clin. Psychiatry 59, 116–122. Ruhe, H.G., Booij, J., Weert, H.C., Reitsma, J.B., Franssen, E.J., Michel, M.C., Schene, A.H., 2009. Evidence why paroxetine dose escalation is not effective in major depressive disorder: a randomized controlled trial with assessment of serotonin transporter occupancy. Neuropsychopharmacology 34, 999–1010. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The
563
mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59 (Suppl. 20), S22–S33. Wernicke, J.F., Dunlop, S.R., Dornseif, B.E., Bosomworth, J.C., Humbert, M., 1988. Low-dose fluoxetine therapy for depression. Psychopharmacol. Bull. 24, 183–188. Zigmond, A.S., Snaith, R.P., 1983. The hospital anxiety and depression scale. Acta Psychiatr. Scand. 67, 361–370.
www.elsevier.com/locate/euroneuro
A placebo-controlled study of three agomelatine dose regimens (10 mg, 25 mg, 25–50 mg) in patients with major depressive disorder$ Sidney H. Kennedya,n, Alla Avedisovab, Natalia Giménez-Montesinosc, Carole Belaïdic, Christian de Bodinatc, On behalf of agomelatine study group1 a
University of Toronto, University Health Network, Canada Serbsky State Scientific Center of Social and Forensic Psychiatry, Russia c Institut de Recherches Internationales Servier (IRIS), 50 rue Carnot, 92284 Suresnes Cedex, France b
Received 3 October 2013; received in revised form 9 January 2014; accepted 11 January 2014
KEYWORDS
Abstract
Agomelatine; Depression; Placebo; Dose regimen
A randomised placebo-controlled “dose relation study” was conducted in 549 patients who met the criteria for major depressive disorder, to evaluate the efficacy and safety of three doses regimens of agomelatine during 6 weeks: low fixed dosage (10 mg/day, n =133), fixed dosage (25 mg/day, n =138) and a flexible dosage with up-titration in case of insufficient improvement
Abbreviations: MDD, major depressive disorder ☆ Trial registration name: Efficacy and safety of 3 dose regimens of agomelatine (10, 25, 25–50 mg) versus placebo given once a day for 6 weeks in out-patients suffering from moderate to severe major depressive disorder. A 6-week randomised, double-blind, placebocontrolled, parallel groups study followed by a double-blind optional 18-week extension period. Trial registration number: EudraCT number: 2009-011238-84. n Correspondence to: FRCPC-University of Toronto EN8-222, 200 Elizabeth Street, Toronto General Hospital, Toronto, ON, Canada M5G 2C4. Tel.: +1 416 340 3888; fax: 1 416 340 4198. E-mail address: [email protected] (S.-n. Kennedy). 1 The agomelatine study CL3-20098-069 group of principal investigators: ARGENTINA: Dr. Ricardo Marcelo Corral; Dr. Mariano Alvarez Caches; Dr. Héctor Fabián Lamaison; Dr. Pedro Rafael Gargoloff; Dr. Christian Lupo ; Dr. Carlos Finkelsztein; Dr. Germán Mauricio Berardo; Dr. Daniel Mosca; Dr. Edith Mirta Serfaty; Dr. Enrique Kuper. BULGARIA: Pr. Vihra Milanova; Pr. Georgi Onchev; Pr. Radka Gaidarova; Pr. Georgi Popov; Dr. Loris Sayan; Dr. Svetlozar Georgiev; Pr. Valentin Akabaliev; Dr. Temenuzhka MatevA. FINLAND: Dr. Antti Ahokas; Dr. Jukka Penttinen; Dr. Marja Pirilä. RUSSIA: Pr. Alla Avedisova; Pr. Elena Grigorieva; Pr. Mikhail Ivanov; Pr. Lala Kasimova; Pr. Sergei Mosolov; Pr. Nikolay Neznanov; Pr. Anatoly Smulevich; Dr. Anna Vasilyeva; Pr. Oleg Shiryaev; Pr. Vladimir Tochilov. SLOVAKIA: Dr. Livia Vavrusova; Dr. Vladimir Garaj; Dr. Nada Kuriackova; Dr. Zuzana Janikova; Dr. Monika Biackova. UKRAINE: Pr. Oleg Chaban; Pr. Volodymyr Abramov; Pr. Olexandr Filts; Dr. Vladyslav Demchenko; Pr. Nataliya Maruta; Dr. Sergiy Mykhnyak; Pr. Oleksandr Napryeyenko; Pr. Valeriy Pidkorytov; Pr. Vitaliy Pishel. http://dx.doi.org/10.1016/j.euroneuro.2014.01.006 0924-977X & 2014 Elsevier B.V. and ECNP. All rights reserved.
554
S.H. Kennedy et al. at week 2 (25–50 mg/day, n =137). At last post-baseline assessment, there were significant and incremental placebo-agomelatine differences on mean HAM-D17 total scores in favour of each agomelatine dose regimen (2.4670.76 points, p=0.001 at 10 mg; 4.71 +0.75 points, po0.0001 at 25 mg and 4.9270.76 points, po0.0001 at 25–50 mg) with statistically significant differences between 25 mg and 25–50 mg dose regimens compared to the 10 mg dose. The response rate according to HAM-D17 was significantly higher in patients taking agomelatine than those taking placebo (difference of 16.1% at 10 mg p=0.005; 25.9% and 27.4% respectively at 25 mg and 25– 50 mg, po0.0001). The benefit of agomelatine was demonstrated in the subgroup of severely depressed patients in the 25 mg and 25–50 mg/day regimens. Consistent clinical response according to CGI variables and better social functioning were found in patients receiving agomelatine. All dose regimens of agomelatine were well tolerated and no unexpected adverse event was reported. This study provides evidence of a dose effect for agomelatine between 10 mg and the therapeutic dose regimen of agomelatine 25–50 mg: the efficacy of the higher dose regimens being more efficacious than the lowest (10 mg) daily dose. The data support a definitive statement regarding the utility of 25 mg as the threshold dose for initiating agomelatine in depressed patients. & 2014 Elsevier B.V. and ECNP. All rights reserved.
1.
Introduction
When selecting a first-line antidepressant agent, considerations include the potential to achieve a greater response with increasing dosages, thereby allowing greater flexibility in dosing. Indeed, increasing the dose of antidepressants seems to be the most straightforward strategy when depressed patients have an insufficient response after adequate time (generally at least 8 weeks) at the initial therapeutic dose (Fredman et al., 2000). However, for most available antidepressants, there are few data from randomised controlled trials to support or refute this approach (Berney, 2005). In the case of SSRI and SNRI antidepressants, it has been difficult to establish a response relationship (Beasley Jr. et al., 1990; Bech et al., 2002; Burke et al., 2002; Dunner and Dunbar, 1992; Fabre et al., 1995; Fabre and Putman III, 1987; Khan et al., 1998; Mallinckrodt et al., 2006; Mendels et al., 1993; Montgomery et al., 1993, 1994; Rudolph et al., 1998; Ruhe et al., 2009; Wernicke et al., 1988). The threshold dose of agomelatine, an antidepressant acting as a MT1 and MT2 receptor agonist and a 5-HT2C receptor antagonist (de Bodinat et al., 2010), is 25 mg/day. This active dose was established in an 8-week double-blind placebo-controlled dose finding study evaluating three dose regimens (1, 5 and 25 mg once a day) (Loo et al., 2002) and several successful placebo controlled or head to head trials have subsequently verified the therapeutic effectiveness of both fixed 25 mg and one step titration 25–50 mg dose regimens (de Bodinat et al., 2010; Hickie and Rogers, 2011; Kennedy and Rizvi, 2010). Following the European Medical Agency requirement at the time of the centralised European Marketing Authorisation, a post-commitment dose relation study was scheduled to ascertain the threshold dose for the therapeutic efficacy of agomelatine in depressed patients and provide evidence for a dose effect of agomelatine. The purpose of this study was to evaluate the efficacy and safety of low dose agomelatine compared to standard doses
of agomelatine in a randomised, placebo controlled trial involving patients with major depressive disorder (MDD). A 10 mg low dose was selected as an intermediate dose between 5 mg (previously assessed in Loo et al. (2002)) and 25 mg and was compared to currently recommended dosages of 25 mg and 25 mg uptitrated to 50 mg for patients with an insufficient improvement after 2 weeks.
2.
Procedures
The study was initiated only after the written Ethics Committee's approval, in accordance with the local regulations in each of the countries, and was conducted in accordance with the principles of Good Clinical Practice and the Declaration of Helsinki. Subjects signed informed consent prior to participating in the trial.
2.1.
Patients
Male or female outpatients aged between 18 and 65 years (inclusive) with a primary diagnosis of MDD of moderate to severe intensity according to DSM-IV-TR (American Psychiatric Association, 2000) and a current major depressive episode (MDE) confirmed by the Mini International Neuropsychiatric Interview (Sheehan et al., 1998) were recruited in 45 centres in 6 countries: Argentina (102 patients), Bulgaria (95 patients), Finland (51 patients), Russia (156 patients), Slovakia (61 patients), and Ukraine (84 patients) between October 2009 and November 2011. Patients had to have a Hamilton Depression Rating Scale 17 items (HAM-D17) total score Z 22, a score of 44 on the item 1 (severity of illness) of the Clinical Global Impression (CGI) scale (Guy, 1976), a Hospital and Depression Anxiety (HAD) (Zigmond and Snaith, 1983) depression subscore of Z 11. Subjects with a decrease of greater than 20% on the HAM-D17 between selection and inclusion were excluded. All patients were required to be physically healthy or to have stabilised somatic illnesses.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder Patients were excluded if they had an ongoing episode of MDD of less than 4 week-duration or with seasonal pattern, psychotic or catatonic features, post-partum onset or onset after stroke within 3 months; if they had resistant depression or no previous response to agomelatine treatment; if they were at risk of suicide; if they had been treated with electroconvulsive therapy (ECT) within the previous 3 months; if they had insight-oriented and structured psychotherapy within the previous 3 months, or lighttherapy within the previous 2 weeks. All types of depression other than MDD according to DSM-IV-TR criteria were excluded as well as current panic disorder, obsessive compulsive disorder, post-traumatic stress disorder, acute stress disorder, acute or chronic psychosis, antisocial, borderline or histrionic personality disorders or any other severe personality disorder (paranoid, schizoid, narcissistic, avoidant, dependent) or alcohol or drug abuse or dependence within the past 12 months. Apart from agomelatine, patients did not receive any other psychotropic treatment during the study, or any treatment prone to interfere with the central nervous system.
2.2.
Treatment allocation
Using an Interactive Response System and according to a balanced (non-adaptive) randomisation with stratification for each centre, patients were assigned to receive agomelatine 10 mg, agomelatine 25 mg, agomelatine 25 mg increased up to 50 mg (one step titration dose regimen), or placebo. For the one step titration regimen, insufficient improvement after 2 weeks led to an increase in the dosage of agomelatine to 50 mg daily, according to a predefined dose adjustment algorithm. Both investigators and subjects remained blind to the up-titration.
2.3.
Measures
The primary outcome measure was the HAM-D17 total score, rated at the selection visit and at weeks 0, 2, 4, and 6. The primary outcome variable was the last post-baseline HAMD17 total score over 6 weeks. Secondary outcome measures were the HAM-D17 response rate (defined as a Z 50% decrease in HAM-D17 total score from baseline), CGI-S score at each visit from selection to week 6 and CGI-I at each visit from week 2 to week 6, CGI-I response rate (defined as CGI-I score = 1 or 2), and Sheehan Disability Scale (SDS) scores at selection and week 6. Safety measures included adverse events reported at each visit, 12-lead ECGs and laboratory tests (biochemistry, haematology) at selection and week 6. A physical examination (blood pressure, heart rate, weight and body mass index) was performed at selection, weeks 0 and 6. If the patient withdrew prematurely, the above-mentioned efficacy and safety measurements were repeated at the time of withdrawal.
2.4.
Statistical analysis
The efficacy analyses were performed on the Full Analysis Set (FAS, randomised patients who took at least one dose of medication, with a value at baseline and at least one post-
555
baseline value for the primary efficacy criterion). The primary analysis examined agomelatine-placebo differences on the last post-baseline value of the HAM-D17 total score over the 6-week period, by using a two-way analysis of covariance (ANCOVA) model on factor treatment (including the 4 treatment groups) with centre (random effect), and baseline HAM-D17 total score as covariates and without interaction. The Hochberg procedure was used to deal with multiplicity issues, providing Hochberg-adjusted p-values. A sensitivity analysis to the method of handling missing values was performed in the FAS. Treatment groups were compared to placebo on the value at week 6, using a mixedeffects model for repeated measures (MMRM), including terms for the fixed effects of treatment, baseline HAM-D17 total score, visit and an interaction term for treatment and visit, and for the random effect of centre. The analysis fitted an unstructured covariance matrix. In the same way as for the main analysis, the Hochberg procedure was also applied. Additional analyses were conducted in the FAS using a χ2 test to assess agomelatine-placebo differences in response to treatment taking into account the last postbaseline value over 6 weeks. The same analyses were also repeated in the planned subset of patients with both HAM-D17 total score Z25 and CGI-S scoreZ 5 at baseline (severely ill patients). Treatment groups were compared to placebo over the 6week period in the FAS for (1) the last post-baseline value of CGI-S score and last value of CGI-I score by using a two-sided Student's t-test for independent samples and a Mann– Whitney test; (2) the response to treatment according to the CGI-I (global improvement score = 1 or 2), using a χ2 test on the last value and (3) SDS work/daily activities, social life and family life scores, number of days lost and number of unproductive days taking into account the mean change from baseline to last post-baseline value using a two-sided Student's t-test (post-hoc analysis). The difference on the primary outcome measures between agomelatine 10 mg versus agomelatine 25 mg and agomelatine 25–50 mg was estimated in the FAS using a two-way ANCOVA model, as for the main analysis but including only the 3 agomelatine treatment groups (post-hoc analysis). For every safety measurement, descriptive statistics were provided by treatment group in the safety set (all included patients who took at least one dose of study medication). Statistical analysis was performed using the SASs software, version 9.1 or 9.2 (SAS Inc.; Cary, North Carolina). The type I error was set at 5% (two-sided tests).
3. 3.1.
Results Patients
Five hundred and forty nine patients were randomised to receive agomelatine 10 mg (133 patients), agomelatine 25 mg (138 patients), a one step titration dosage of agomelatine 25–50 mg (137 patients) or placebo (141 patients). A total of 505 patients completed the 6-week treatment period (91.9%). Reasons for premature withdrawal were mainly lack of efficacy or non-medical reason (Figure 1). There were no clinically relevant differences
556
S.H. Kennedy et al.
Screened N = 683
Not Randomized Not fulfilled study entry criteria Adverse events Withdrew consent
N = 134 n=116 n = 10 n=8
Randomized N = 549
Agomelatine 10 mg N = 133
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
n=123 n=10 n=1 n=2 n=5 n=2
Agomelatine 25 mg fixed N = 138
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
Figure 1
n=130 n=8 n=1 n=1 n=3 n=3
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
n=126 n=11 n=1 n=1 n=5 n=4
Placebo N = 141
Completed at week 6 Withdrawn on weeks0-6 Adverse event Protocol deviation Lack of efficacy Non-medical reason
n=126 n=15 n=0 n=0 n=9 n=6
Disposition of patients.
between the groups for demographic and clinical characteristics (Table 1). In the group of patients who were eligible for up-titration, 49 out of 134 patients (36.6%) had a dose increase of agomelatine from 25 mg to 50 mg after 2 weeks.
3.2.
Agomelatine 25 or 25-50 mg N = 137
Efficacy
3.2.1. Whole study population The mean HAM-D17 total score decreased from baseline to week 6 in all groups (Figure 2). Agomelatine was associated with a clinically relevant decrease in symptoms at the last post-baseline value (Table 2). The placebo-agomelatine differences ranged from 2.4670.76 points (p= 0.001) in the 10 mg group, to 4.7170.75 (po0.0001) in the 25 mg group and 4.9270.76 points (po0.0001) in the group of patients with potential one step titration 25–50 mg dosage regimen. Results from the MMRM sensitivity analysis were consistent with the main analysis. The mean HAM-D17 total score at the last post-baseline assessment was not statistically different between the agomelatine 25 mg and 25–50 mg groups (E(SE)= 0.21 (0.67), 95% CI= [ 1.12; 1.54]). The mean last postbaseline HAM-D17 total score in the agomelatine 10 mg group was statistically significantly smaller than in both agomelatine 25 mg (E(SE) = 2.24 (0.72), 95% CI= [0.82; 3.65]) and agomelatine 25–50 mg groups (E(SE)= 2.44 (0.72), 95% CI= [1.02; 3.86], post-hoc analysis). Regarding secondary efficacy outcomes, the response rate according to HAM-D17 was significantly higher in patients taking agomelatine than those taking placebo, with a clinically relevant difference from placebo of 16.1% on agomelatine 10 mg (p= 0.005), 25.9% on agomelatine 25 mg (po0.0001), and 27.4% on agomelatine 25–50 mg (po0.0001) (Table 2). The differences of the mean CGI-S
and CGI-I scores between the 3 agomelatine dose regimens and placebo were statistically significant at the last (postbaseline) value with the Student's t-test (except for a trend to statistical significance for agomelatine 10 mg (p= 0.076) for the mean CGI-S score) and with the Mann–Whitney test (Table 2). The percentage of responders according to CGI-I was significantly higher in the three agomelatine groups (from 52.3% to 71.7%) than in the placebo group (36.9%) (Table 2). Over the first 6 weeks of treatment, mean decreases in the three SDS sub-scales; the mean decrease in number of lost work days, and number of underproductive days at the last post-baseline assessment were significantly greater in both fixed 25 mg and one step titration 25–50 mg agomelatine groups, than in the placebo group. In contrast, in the agomelatine 10 mg group, the effects on “Work and activity” score and the number of underproductive days, were not significantly different from those obtained in the placebo group (Table 3).
3.2.2. Severely ill patients The mean HAM-D17 total score decreased from baseline to week 6 in all groups (Figure 3). In the subset of severely depressed patients with baseline HAM-D17 total score Z25 and CGI-S Z 5, at the last post-baseline value, the placeboagomelatine differences on the HAM-D17 total score did not achieve statistical significance at the 10 mg regimen dosage (1.5771.16, p =0.176) but was highly significant at fixed 25 mg (5.3871.16, po0.0001) and at the one step titration 25–50 mg (5.5371.20, po0.0001) dosages (Table 4). Results from the MMRM sensitivity analysis were consistent with the main analysis. The response rate according to HAM-D17 was significantly higher in patients on agomelatine 25 mg fixed and one step titration 25–50 mg dosage than on placebo,
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder
Table 1
557
Demographics and baseline characteristics (randomised set).
Characteristics
Agomelatine
Age, mean7SD [range] Gender, female n (%) DSM-IV MDE single, n (%) MDE recurrent, n (%) Current MDE duration before inclusion (months), median HAM-D-17 total score, mean7SD CGI-S score, mean7SD
10 mg (N =133)
25 mg fixed (N =138)
25–50 mg (N =137)
Placebo (N =141)
45.4712.6 [18;65] 105 (79.0)
46.5711.9 [20;65] 90 (65.2)
43.1712.6 [19;65] 98 (71.5)
45.0713.0 [18;65] 108 (76.6)
33 (24.8) 100 (75.2) 2.50
30 (21.7) 108 (78.3) 2.50
34 (24.8) 103 (75.2) 2.50
35 (24.8) 106 (75.2) 2.50
27.272.7 4.770.6
26.772.9 4.770.6
26.772.9 4.670.6
26.672.6 4.670.7
Abbreviations: MDE=major depressive episode; DSM-IV=diagnostic and statistical manual of mental disorders (fourth edition, Text Revision); HAM-D-17=Hamilton depression rating scale 17 items; CGI-S=Clinical Global Impression Severity of illness score.
30
RANDOMISATION TREATMENT
29
Agomelatine 10 mg Agomelatine 25 or 25-50 mg
Agomelatine 25 mg fixed Placebo
28 27 26 25 24
MEAN VALUE
23 22 21 20 19 18 17 16 15 14 13 12 11 10 W000
W002
W004
W006
LAST POST-BASELINE (ON W000-W006)
VISIT
Figure 2
Course of the HAM-D score over the time of 6 weeks for all four treatment-groups in the whole study population.
with a clinically relevant placebo-agomelatine difference of 25.0% and 25.8%, respectively (p= 0.002 in each group respectively).
3.3.
Tolerability
During the 6-week study period, in the safety set, the percentage of patients with at least one emergent adverse event was higher in the 3 agomelatine groups (10 mg [24.2%], 25 mg [27.5%] and one step titration 25–50 mg [35.3%]) than in the placebo group (18.4%). Headache and
nausea were the most frequent emergent adverse events (EAEs) reported by patients in agomelatine groups (Table 5). The majority of adverse events were rated as mild or moderate. In all, 3 patients had severe emergent adverse events (1 patient each in the agomelatine 10 mg, 25 mg fixed and placebo groups). The percentage of patients who experienced at least one emergent adverse event leading to study treatment discontinuation was 0.8% (1 patient) in the agomelatine 10 mg group, 0.7% (1 patient) in the agomelatine 25 mg fixed group, 1.5% (2 patients) in the agomelatine one step titration 25–50 mg group, and 0.7% (1 patient) in the placebo group. Serious adverse events reported in one
558
Table 2
S.H. Kennedy et al.
Primary and key secondary efficacy outcomes over 6 weeks (Full Analysis Set). Agomelatine 10 mg (N =132)
25 mg fixed (N =138)
25–50 mg (N =136)
Placebo (N =141)
HAM-D-17 total score (a) n Baseline, mean7SD Last post-baseline value, mean7SD Difference from placebo (SE)a 95% CI p-Value b
132 27.372.6 16.577.0 2.46 (0.76) [0.96;3.96] 0.001 (III)
138 26.772.9 14.076.6 4.71 (0.75) [3.23;6.19] o0.0001 (II)
136 26.772.9 13.976.3 4.92 (0.76) [3.43;6.40] o0.0001 (I)
141 26.672.6 18.777.3
MMRM (b) n at week 6 Value at week 6, mean7SD Difference from placebo (SE)a 95% CI p-Valueb
126 16.176.8 2.83 (0.77) [1.32;4.34] o0.001 (III)
131 13.776.4 4.81 (0.76) [3.32;6.30] o0.0001 (II)
130 13.375.7 5.08 (0.76) [3.58;6.57] o0.0001 (I)
129 18.077.0
HAM-D-17 response rate (c) n Last post-baseline value, n (%) Difference from placebo (SE)c 95% CI p-Valued
132 54 (40.9) 16.1 (5.6) [ 27.1; 5.1] 0.005
138 70 (50.7) 25.9 (5.6) [ 36.9; 14.9] o0.0001
136 71 (52.2) 27.4 (5.6) [ 38.4; 16.4] o0.0001
141 35 (24.8)
CGI-S score (d) n Last post-baseline value, mean7SD Difference from placebo (SE)e 95% CI p-Valuef p-Valueg
132 3.571.1 0.25 (0.14) [ 0.03;0.52] 0.076 0.044
138 3.171.1 0.67 (0.13) [0.41;0.92] o0.0001 o0.0001
136 3.171.0 0.65 (0.13) [0.39;0.91] o0.0001 o0.0001
141 3.771.1
CGI-I score (d) n Last value, mean7SD Difference from placebo (SE) 95% CI p-Valuef p-Valueg
132 2.671.0 0.33 (0.13) [0.07;0.59] 0.013 0.009
138 2.271.0 0.76 (0.13) [0.51;1.01] o0.0001 o0.0001
136 2.270.9 0.77 (0.12) [0.52;1.01] o0.0001 o0.0001
141 2.971.1
CGI-I response rate (c) n Last value, n (%) Difference from placebo (SE) 95% CI p-Value d
132 69 (52.3) 15.4 (6.0) [ 27.1; 3.7] 0.011
138 99 (71.7) 34.9 (5.6) [ 45.8; 23.9] o0.0001
136 97 (71.3) 34.4 (5.6) [ 45.5; 23.4] o0.0001
141 52 (36.9)
(a) Analysis of covariance model on factor treatment with baseline HAM-D-17 total score and centre (random effect) as covariates; (b) mixed-effects model with repeated measures including terms for effects of treatment, baseline HAM-D-17 total score, centre (random effect), visit and an interaction term treatment visit; (c) χ2 test; and (d) two-sided Student's t-test for independent samples/ Mann–Whitney test. CGI-S=Clinical Global Impression Severity of illness score; CGI-I =clinical global impression global improvement score; HAM-D17=Hamilton depression rating scale 17 items a Difference between adjusted treatment group means (placebo minus agomelatine). b Two-sided Hochberg-adjusted p-value with (I)/(II)/(III) corresponding to the ordered unadjusted p-values from the most significant to the least significant one. c Difference between percentages (Placebo minus Agomelatine). d Two-sided p-value. e Difference between treatment group means (Placebo minus Agomelatine). f Student's t-test: two-sided p-value. g Mann–Whitney test: two-sided p-value.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder
Table 3
559
Mean SDS change from baseline to last post-baseline assessment over 6 weeks (Full Analysis Set). Agomelatine 10 mg (N =132)
25 mg fixed (N =138)
Work and activitya n Mean7SD Difference from placebo (SE)b 95% CI p-Value c
98 2.172.2 0.48 (0.30) [ 0.11;1.08] 0.112
102 2.972.6 1.32 (0.33) [0.68;1.97] o0.0001
103 2.972.2 1.36 (0.29) [0.79;1.94] o0.0001
102 1.672.0
Social life n Mean7SD Difference from placebo (SE) 95% CI p-Valuec
127 2.572.4 0.78 (0.28) [0.23;1.33] 0.005
127 3.072.6 1.25 (0.29) [0.67;1.83] o0.0001
128 3.172.5 1.34 (0.29) [0.78;1.90] o0.0001
136 1.772.2
Family life and home responsibilities n 127 Mean7SD 2.572.4 Difference from placebo (SE) 0.73 (0.29) 95% CI [0.16;1.30] p-Valuec 0.012
127 3.072.6 1.28 (0.30) [0.68;1.88] o0.0001
128 3.172.4 1.35 (0.29) [0.78;1.93] o0.0001
135 1.772.3
Number of days lost n Mean7SD Difference from placebo (SE) 95% CI p-Valuec
120 1.472.3 0.54 (0.27) [0.01;1.07] 0.046
120 1.972.6 0.98 (0.28) [0.42;1.54] o0.001
127 0.971.9
121 2.472.5 1.06 (0.32) [0.42; 1.69] 0.001
121 2.872.6 1.39 (0.32) [0.75; 2.03] o0.0001
126 1.472.5
123 1.472.2 0.52 (0.26) [0.02;1.03] 0.044
Number of underproductive days n 126 Mean7SD 1.872.8 Difference from placebo (SE) 0.40 (0.33) 95% CI [ 0.26; 1.05] 0.236 p-Valuec
25–50 mg (N =136)
Placebo (N =141)
a
Only values of patients having worked/studied during the past week. Difference between treatment group means (Placebo minus Agomelatine). c Two-sided Student's t-test for independent samples. b
patient (0.7%) receiving agomelatine 25 mg (depression, insomnia, anxiety, restlessness and decreased appetite) and one patient (0.7%) in the placebo group (haemorrhagic enterocolitis) were considered as unrelated to treatment by the investigator. No death occurred during the study. EAEs related to study treatment were reported more often by patients taking agomelatine (10.6%, 13.8%, and 15.4% in the agomelatine 10 mg, 25 mg fixed, and one step titration 25–50 mg groups, respectively) than by those taking placebo (6.4%). There were no clinically relevant between-group differences or changes from baseline to the last value during treatment for weight and vital signs, blood pressures and heart rate, biochemical and haematological parameters. No clinically relevant ECG abnormalities were recorded. During the 6-week treatment period of the study, no
emergent potentially clinically significant abnormal transaminase levels (43 ULN) were observed (Table 6).
4.
Discussion
This placebo-controlled study demonstrates the clinical efficacy and safety of three dose regimens of agomelatine given once a day for 6 weeks in out-patients suffering from moderate to severe major depressive disorder. Each dose regimen arm separated significantly from placebo with relevant differences of 2.46, 4.71 and 4.92 points on the final HAM-D17 total score respectively for the 10 mg, 25 mg, and 25–50 mg agomelatine groups. The antidepressant efficacy is supported by consistent differences in rates of response compared to placebo; 16% agomelatine-placebo
560
S.H. Kennedy et al. 30 RANDOMISATION TREATMENT
29
Agomelatine 10 mg Agomelatine 25 or 25-50 mg
Agomelatine 25 mg fixed Placebo
28 27 26 25 24
MEAN VALUE
23 22 21 20 19 18 17 16 15 14 13 12 11 10 W000
W002
W004
W006
LAST POST-BASELINE (ON W000-W006)
VISIT
Figure 3 Course of the HAM-D score over the time of 6 weeks for all four treatment-groups in the subset of severely ill patients. Table 4 Efficacy outcomes over 6 weeks in the more severely depressed patients at inclusion (Sub-FAS with baseline HAM-D17 total score Z25 and CGI-S Z5). Agomelatine 10 mg (N =132)
25 mg fixed (N =138)
25 50 mg (N =136)
Placebo (N =141)
HAM-D-17 total score (a) n Baseline, mean7SD Last post-baseline value, mean7SD Difference from placebo (SE)a 95% CI p-Valueb
73 28.572.2 18.377.0 1.57 (1.16) [ 0.71; 3.85] 0.176 (III)
70 28.272.9 14.277.0 5.38 (1.16) [3.10; 7.67] o0.0001 (I)
63 28.772.8 14.376.6 5.53 (1.20) [3.18; 7.89] o0.0001 (II)
68 28.272.6 19.677.7
MMRM (b) n at week 6 Value at week 6, mean7SD Difference from placebo (SE)a 95% CI p-Valueb
71 18.176.9 1.72 (1.14) [ 0.53; 3.97] 0.133 (III)
68 14.277.0 5.27 (1.15) [3.01;7.53] o0.0001 (II)
59 13.475.8 5.74 (1.19) [3.40;8.07] o0.0001 (I)
61 18.777.5
HAM-D-17 response rate (c) n Last post-baseline value, n (%) Difference from placebo (SE)c 95% CI p-Valued
73 26 (35.6) 10.6 (7.7) [-25.7;4.4] 0.171
70 35 (50.0) 25.0 (8.0) [ 40.6; 9.4] 0.002
63 32 (50.8) 25.8 (8.2) [ 41.9; 9.7] 0.002
68 17 (25.0)
(a) Analysis of covariance model on factor treatment with baseline HAM-D-17 total score and centre (random effect) as covariates; (b) mixed-effects model with repeated measures including terms for effects of treatment, baseline HAM-D-17 total score, centre (random effect), visit and an interaction term treatment visit; (c) χ2 test. a Difference between adjusted treatment group means (Placebo minus Agomelatine). b Two-sided Hochberg-adjusted p-value with (I)/(II)/(III) corresponding to the ordered unadjusted p-values from the most significant to the least significant one. c Difference between percentages (Placebo minus Agomelatine). d Two-sided p-value.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder
561
Table 5 Emergent adverse events during the double-blind 6-week treatment period in at least 2% of any agomelatinetreated patients (Safety Set). Agomelatine
Headache Nausea Nasopharyngitis Diarrhoea Dizziness Somnolence Constipation Dry mouth
10 mg (N= 132)
25 mg fixed (N =138)
25–50 mg (N =136)
Placebo (N= 141)
11 (8.3) 3 (2.3) – 1 (0.8) – 3 (2.3) – 1 (0.8)
6 7 – 2 2 1 3 3
10 (7.4) 5 (3.7) 4 (2.9) 4 (2.9) 4 (2.9) 3 (2.2) 1 (0.7) 1 (0.7)
7 2 4 – – 2 1 –
(4.3) (5.1) (1.4) (1.4) (0.7) (2.2) (2.2)
(5.0) (1.4) (2.8)
(1.4) (0.7)
Number of patients (%).
Table 6 ALT/AST parameters during the double-blind 6-week treatment period. Mean changes from baseline to last value (Safety Set). Agomelatine
ASAT (IU/L) ALAT (IU/L)
n Mean7SD n Mean7SD
10 mg (N =132)
25 mg fixed (N =138)
25–50 mg (N =136)
Placebo (N =141)
125 0.676.2 125 0.379.7
129 0.177.4 129 1.3711.3
128 0.475.8 128 0.1710.0
134 0.376.5 134 0.3710.9
difference on agomelatine 10 mg; 26% and 27% on agomelatine 25 mg and 25–50 mg respectively and similar difference on CGI variables. This study also confirms a dose effect of agomelatine, with both superiority of the 10 mg dose over placebo and superiority of agomelatine 25 mg and 25–50 mg over both agomelatine 10 mg and placebo. Efficacy of the two higher dose regimens (25 mg and 25–50 mg) of agomelatine compared to placebo is further substantiated in the predefined ‘severe depression’ subpopulation, who displayed a 5 point agomelatine-placebo difference at endpoint according to HAM-D17 total score and a 25% difference from placebo in clinical response. Thus, those patients likely to be most impaired by their depression and most likely to receive antidepressant treatment in routine practice need a minimum dosage of 25 mg daily. There is a bilateral relationship between depressive symptoms, reduced social capabilities and functional impairment. Based on SDS findings, agomelatine helps to improve symptom-related functional impairment in patients over 6 weeks, and those improvements were noted on all domains for the fixed dose 25 mg, and the one step titration 25–50 mg dose regimen. This translates into better social functioning and quality of life for patients, as has been previously reported at both fixed 25 mg and one step titration 25– 50 mg dosages (Goodwin, 2009; Heun et al., in press). In contrast, the absence of significant effects on “Work and activity” score and ‘number of underproductive days’ in the agomelatine 10 mg group provides further evidence that the
lowest tested dose displays lower efficacy when compared to agomelatine 25 mg with or without up-titration to 50 mg. Finally, the three dose regimens of agomelatine were well tolerated during the 6-week treatment period. Compared to the placebo group, higher incidences of emergent adverse events were reported in the 3 agomelatine groups, and this is compatible with the observation that side-effects commonly increase with increased dosing. However, no unexpected adverse event occurred, supporting previous reports on the side effect profile of agomelatine (de Bodinat et al., 2010; Hickie and Rogers, 2011; Kennedy and Rizvi, 2010), and there were no specific concerns about liver enzymes on the short-term in this study. These findings are noteworthy as they represent the first demonstration of a dose response effect of an antidepressant medication, with statistically significant differences between doses. To date, when reviewing the literature, only equivocal findings were obtained regarding a dose–response relationship for all currently available antidepressants, notably among the SSRIs, fluoxetine (Fabre and Putman III, 1987; Wernicke et al., 1988; Beasley Jr. et al., 1990), citalopram (Bech et al., 2002; Montgomery et al., 1994; Montgomery et al., 1993), escitalopram (Burke et al., 2002), paroxetine (Dunner and Dunbar, 1992; Ruhe et al., 2009) and sertraline (Fabre et al., 1995). Neither was a dose-relationship demonstrated for venlafaxine (Khan et al., 1998; Mendels et al., 1993; Rudolph et al., 1998) or duloxetine (Mallinckrodt et al., 2006). The dose effect for agomelatine between 10 mg and therapeutic dose regimen of agomelatine 25–50 mg complements
562 and extends the original dose finding study by Loo et al. (2002), in which 25 mg proved the most efficacious compared to 1 mg and 5 mg and represents the threshold dose for initiating agomelatine in a secure way for the patients. The clinically relevant findings obtained with the one step titration 25–50 mg dose also confirm the utility of 50 mg for some patients. In routine clinical practice, when no early treatment improvement is observed, the clinician classically increases the dose of treatment before considering a change in therapeutic management.
Role of the funding source This study was sponsored by Servier (Suresnes, France). The sponsor of the study participated in study design, data collection, data analysis, data interpretation, and writing of the report. The authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Contributors The sponsor of the study participated in study design, data collection, data analysis, data interpretation, and writing of the report. The authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis
Conflict of interest Sidney Kennedy has acted as advisor/consultant or speaker for AA Pharma, Abbott Laboratories, AstraZeneca, Bristol-Myers Squibb, Eli Lilly, Elsevier, Forest Laboratories, Glia Communications, Indigene Lifesystems Pvt. Ltd., Lundbeck, Lundbeck Institute, Pfizer, Servier, Spimaco, Symbiote Medical Communications Inc. and Spirant Communications. He has received grant support from Bristol-Myers Squibb, Brain Cells Inc., Canadian Institutes of Health Research, Clera Inc., Eli Lilly, Lundbeck, Ontario Brain Institute, Pfizer, Servier and St. Jude Medical Inc. Alla Avedisova has received grant/research support from GlaxoSmithKline, Lundbeck, Sanofi and Servier; has served as a consultant or on advisory boards for Janssen and Servier Natalia Giménez-Montesinos, Carole Belaïdi and Christian de Bodinat are employees at Servier.
Ackowledgement None.
References American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. American Psychiatric Association, Washington, DC. Beasley Jr., C.M., Bosomworth, J.C., Wernicke, J.F., 1990. Fluoxetine: relationships among dose, response, adverse events, and plasma concentrations in the treatment of depression. Psychopharmacol. Bull. 26, 18–24. Bech, P., Tanghoj, P., Andersen, H.F., Overo, K., 2002. Citalopram dose–response revisited using an alternative psychometric approach to evaluate clinical effects of four fixed citalopram
S.H. Kennedy et al. doses compared to placebo in patients with major depression. Psychopharmacology (Berl.) 163, 20–25. Berney, P., 2005. Dose–response relationship of recent antidepressants in the short-term treatment of depression. Dialog. Clin. Neurosci. 7, 249–262. Burke, W.J., Gergel, I., Bose, A., 2002. Fixed-dose trial of the single isomer SSRI escitalopram in depressed outpatients. J. Clin. Psychiatry 63, 331–336. de Bodinat, C., Guardiola-Lemaitre, B., Mocaer, E., Renard, P., Munoz, C., Millan, M.J., 2010. Agomelatine, the first melatonergic antidepressant: discovery, characterization and development. Nat. Rev. Drug Discov. 9, 628–642. Dunner, D.L., Dunbar, G.C., 1992. Optimal dose regimen for paroxetine. J. Clin. Psychiatry 53 (Suppl.), S21–S26. Fabre, L.F., Abuzzahab, F.S., Amin, M., Claghorn, J.L., Mendels, J., Petrie, W.M., Dube, S., Small, J.G., 1995. Sertraline safety and efficacy in major depression: a double-blind fixed-dose comparison with placebo. Biol. Psychiatry 38, 592–602. Fabre, L.F., Putman III, H.P., 1987. A fixed-dose clinical trial of fluoxetine in outpatients with major depression. J. Clin. Psychiatry 48, 406–408. Fredman, S.J., Fava, M., Kienke, A.S., White, C.N., Nierenberg, A.A., Rosenbaum, J.F., 2000. Partial response, nonresponse, and relapse with selective serotonin reuptake inhibitors in major depression: a survey of current "next-step" practices. J. Clin. Psychiatry 61 (6), 403–408. Goodwin, G.M., 2009. Clinical studies on the efficacy of agomelatine on depressive symptoms. CNS Drugs 23 (Suppl 2), S35–S39. Guy, W., 1976. ECDEU Assessment Manual for Psychopharmacology. US Department of Health, Education and Welfare Publication ADM 76-338. National Institute of Mental Health, Rockville, MD217–222. Heun, R., Ahokas, A., Boyer, P., Giménez-Montesinos, N., PontesSoares, F., Olivier, V., 2014. The efficacy of agomelatine in elderly patients with major recurrent depressive disorder: a placebo controlled study. J. Clin. Psychiatry. Hickie, I.B., Rogers, N.L., 2011. Novel melatonin-based therapies: potential advances in the treatment of major depression. Lancet 378, 621–631. Kennedy, S.H., Rizvi, S.J., 2010. Agomelatine in the treatment of major depressive disorder: potential for clinical effectiveness. CNS Drugs 24, 479–499. Khan, A., Upton, G.V., Rudolph, R.L., Entsuah, R., Leventer, S.M., 1998. The use of venlafaxine in the treatment of major depression and major depression associated with anxiety: a dose–response study. Venlafaxine investigator study group. J. Clin. Psychopharmacol. 18, 19–25. Loo, H., Hale, A., D'haenen, H, 2002. Determination of the dose of agomelatine, a melatoninergic agonist and selective 5-ht2c antagonist, in the treatment of major depressive disorder: a placebo-controlled dose range study. Int. Clin. Psychopharmacol. 17, 239–247. Mallinckrodt, C.H., Prakash, A., Andorn, A.C., Watkin, J.G., Wohlreich, M.M., 2006. Duloxetine for the treatment of major depressive disorder: a closer look at efficacy and safety data across the approved dose range. J. Psychiatr. Res. 40, 337–348. Mendels, J., Johnston, R., Mattes, J., Riesenberg, R., 1993. Efficacy and safety of b.i.d. doses of venlafaxine in a dose–response study. Psychopharmacol. Bull. 29, 169–174. Montgomery, S.A., Pedersen, V., Tanghoj, P., Rasmussen, C., Rioux, P., 1994. The optimal dosing regimen for citalopram—a metaanalysis of nine placebo-controlled studies. Int. Clin. Psychopharmacol. 9 (Suppl. 1), S35–S40. Montgomery, S.A., Rasmussen, J.G., Tanghoj, P., 1993. A 24-week study of 20 mg citalopram, 40 mg citalopram, and placebo in the prevention of relapse of major depression. Int. Clin. Psychopharmacol. 8, 181–188.
Placebo-controlled trial of 3 dose regimens of agomelatine in major depressive disorder Rudolph, R.L., Fabre, L.F., Feighner, J.P., Rickels, K., Entsuah, R., Derivan, A.T., 1998. A randomized, placebo-controlled, dose– response trial of venlafaxine hydrochloride in the treatment of major depression. J. Clin. Psychiatry 59, 116–122. Ruhe, H.G., Booij, J., Weert, H.C., Reitsma, J.B., Franssen, E.J., Michel, M.C., Schene, A.H., 2009. Evidence why paroxetine dose escalation is not effective in major depressive disorder: a randomized controlled trial with assessment of serotonin transporter occupancy. Neuropsychopharmacology 34, 999–1010. Sheehan, D.V., Lecrubier, Y., Sheehan, K.H., Amorim, P., Janavs, J., Weiller, E., Hergueta, T., Baker, R., Dunbar, G.C., 1998. The
563
mini-international neuropsychiatric interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J. Clin. Psychiatry 59 (Suppl. 20), S22–S33. Wernicke, J.F., Dunlop, S.R., Dornseif, B.E., Bosomworth, J.C., Humbert, M., 1988. Low-dose fluoxetine therapy for depression. Psychopharmacol. Bull. 24, 183–188. Zigmond, A.S., Snaith, R.P., 1983. The hospital anxiety and depression scale. Acta Psychiatr. Scand. 67, 361–370.
