Cns drugs 2006; 20 (12): 981-992

LEADING ARTICLE
2006 Adis Data Information BV. All rights reserved.
Agomelatine
A Preliminary Review of a New Antidepressant
Michael Zupancic and Christian Guilleminault Stanford University Sleep Medicine Program, Stanford, California, USA Abstract
Agomelatine is a new antidepressant that is a potent agonist of melatonin receptors and an antagonist of the serotonin 5-HT2C receptor subtype. It is inlate-phase trials for the treatment of major depressive disorder (MDD).
Symptoms of depression significantly improved with agomelatine compared with placebo in large placebo-controlled trials, and agomelatine appears to be asefficacious in treating MDD as other antidepressants but with fewer adverseeffects. Agomelatine appears to improve sleep quality and ease of falling asleep,as measured subjectively in depressed patients. Polysomnographic studies haveshown that agomelatine decreases sleep latency, decreases wake after sleep onset(WASO), and improves sleep stability as measured by changes in the cyclicalternating pattern.
Agomelatine is generally well tolerated in patients with MDD; in clinical trials, adverse events were generally mild to moderate in nature, with an overallfrequency close to that of placebo. Discontinuation of agomelatine because ofadverse effects occurred at a similar rate to placebo.
Major depressive disorder (MDD) is a common nificant discontinuation symptoms that may ulti- disorder, with a lifetime prevalence of 16.6%, which Some symptoms of depression are thought to is associated with significant morbidity and mortali- be related to the disorganisation of homeostatic ty.[1] Although many individuals experience the dis- rhythms. Many patients with depression have a order, only a small proportion of patients with MDD blunted circadian rhythm, as demonstrated by ab- present for treatment. Older antidepressants such as normal 24-hour temperature curves and melatonin the TCAs, although effective, have significant and levels.[2-4] In addition, 40–60% of outpatients with sometimes life-threatening adverse effects that limit MDD report sleep disturbances, i.e. early morning their use. The newer antidepressants such as the awakenings and difficulties initiating and maintain- SSRIs tend to be better tolerated, but still have ing sleep.[5] A new antidepressant would be advanta- significant adverse effects, e.g. sexual dysfunction geous if it were effective, had fewer adverse effects and nausea, which may limit compliance. Addition- than currently available agents, and helped reor- ally, many antidepressants are associated with sig- ganise the body’s internal rhythms.
Agomelatine is a unique antidepressant that is receptors to frontocortical dopaminergic and adren- currently in late-phase trials for MDD. It is a potent agonist of melatonin receptors and an antagonist ofthe serotonin 5-HT2C receptor subtype. This review 1.3 Effects in Animal Models of Depression summarises the pharmacology of agomelatine and the data on its efficacy and tolerability in the treat- Four published studies have evaluated the effect of agomelatine in animal models of depression,[10-13]and several studies have been published evaluating 1. Pharmacodynamic Properties
the effects of the drug in animal models of anxie-ty.[14,15] Papp et al.[10] evaluated the effects that agomela- tine had on chronic mild stress, a rat model of Agomelatine is a potent agonist of melatonin depression. This study showed that administration MT1 and MT2 receptors (inhibition constant [ki] = of agomelatine at 10 or 50 mg/kg for 7 weeks 6.15 × 10–11 and 2.68 × 10–10 mol/L, respectively).
counteracted the stress-induced decrease in sucrose It is also an antagonist of the serotonin 5-HT2C consumption in a dose-dependent manner. This ef- receptor subtype (concentration that inhibits binding fect was noted whether agomelatine was adminis- of a ligand by 50% [IC50] = 2.7 × 10–7 mol/L), tered in the morning or evening. The involvement of which is concentrated in the frontal cortex, hippo- melatonin receptors in the antidepressant effect of agomelatine was suggested by the fact that concomi- Agomelatine displays low affinity at native (rat)/ tant administration of an MT1/MT2 receptor antago- cloned human 5-HT2A (ki <5.0/5.3 mol/L) and nist (S-22153) in the evening diminished the effects 5-HT1A (ki <5.0/5.2 mol/L) receptors, and negligi- of agomelatine on rodent sucrose consumption.
ble (ki <5.0 mol/L) affinity for other serotonin re- However, agomelatine still showed antidepressant ceptor subtypes.[7] In binding studies of cloned activity when concomitantly administered with a human MT1 and MT2 receptors, agomelatine has melatonin receptor antagonist in the morning, sug- similar affinity to that of melatonin (ki = 8.52 × gesting its antidepressant qualities are mediated 10–11 and 2.63 × 10–10 mol/L, respectively).[7-9] through both its agonist action at MT1/MT2 recep-tors and its antagonist action at 5-HT2C receptors, as shown by animal data.[7] Additionally, the antide- In vivo data indicate that agomelatine enhanced pressant-like activity of agomelatine was similar the levels of dopamine in the frontal cortex of freely to the other antidepressants used in this study,imipramine and fluoxetine.[10] moving rats, whereas these levels were unaffected inthe nucleus accumbens and striatum. In vivo data Millan et al.[14] published a paper evaluating the show that extracellular levels of noradrenaline effects of agomelatine on several rodent models of (norepinephrine) in the frontal cortex of rats were anxiety. The social interaction test consists of plac- dose-dependently enhanced by agomelatine in par- ing two rats together in a novel environment and allel with acceleration in the firing rate of adrenergic measuring the active contact between the animals.
cell bodies in the locus ceruleus. These increases in In this test, doses of agomelatine of 2.5 and 10.0 mg/ noradrenaline and dopamine levels are likely to re- kg elicited a significant increase in the time devoted flect blockade of the inhibitory input of 5-HT2C to active socialisation in pairs of animals, indicating 2006 Adis Data Information BV. All rights reserved.
an anxiolytic effect. The Vogel conflict test is a test hippocampal formation may be a factor in the devel- where water-deprived rats are exposed to electri- opment of depression and is the basis for the cal shocks while drinking water. Agomelatine neuroplasticity hypothesis of major depression.[21] (2.5–80.0 mg/kg) elicited an increase in punished A recent study published by Banasr et al.[22] has responses, suggesting that it alleviates anxiety. In shown that chronic treatment with agomelatine (40 these two models of anxiety, agomelatine was found mg/kg for 3 weeks) increased cell proliferation and to have dose-dependent anxiolytic activities, with neurogenesis in the ventral dentate gyrus, a region effects comparable to the benzodiazepine cloraze- implicated in the response to anxiety and emotion, in nonstressed rats. Chronic treatment with agomela- Agomelatine also alleviated anxiety in the plus- tine also increased the survival of these newly maze test, another rodent model of anxiety.[14] This formed cells. These findings suggest that the antide- test consists of placing rodents in the centre of a pressant and anxiolytic effects of agomelatine may maze-like structure and recording how frequently be partially due to effects on the ventral dentate they explore different arms of the maze. Agomela- tine evoked a modest increase of exploration athigher doses of 40 and 80 mg/kg, but with a less 1.5 Effects on Circadian Rhythms in Animals robust effect than clorazepate.[14] Unlike cloraze-pate, agomelatine did not decrease levels of seroto-nin and noradrenaline in the hippocampus and fron- Because of the agonistic activity of agomelatine tal cortex of rats.[14] These data suggest that the at MT1/MT2 receptors, several studies using animal anxiolytic effects of agomelatine are mediated models have evaluated the effects of the drug on the through a different mechanism than those of Redman et al.[23] studied the effect of agomela- tine in a rat model of phase advancement and foundthat it entrained the circadian rhythm similar to melatonin. Martinet et al.[24] demonstrated thatagomelatine entrained the circadian rhythm of free- Several clinical studies have shown that patients running rats to a similar degree to melatonin. Van with stress-induced depression have reduced hippo- Reeth et al.[25] showed that the drug enhanced the campal volume.[16,17] This reduction is thought to ability of old hamsters to phase shift compared with occur because of the loss or atrophy of glial and controls. Agomelatine has also been found to ad- neuronal cells, potentially caused by increases in vance the circadian rhythm in a rat model of the corticosteroids and excitatory amino acids sec- ondary to stress. Although this has been demonstrat-ed in animals, in humans it is currently still under Overall, these animal studies demonstrate that agomelatine is able to resynchronise a disrupted The dentate gyrus of the hippocampal formation circadian rhythm and has circadian phase-advance- is a site of continuous neurogenesis during adult life ment properties. The chronobiotic effect of agome- in humans and is involved in some form of learning latine is mediated through its effects on the and memory.[20] In animal models, treatment with suprachiasmic nucleus via the M1 and M2 recep- certain antidepressants can prevent or reverse some tors,[27] and the extent of the effect is similar to that of these structural alterations. Remodelling of the 2006 Adis Data Information BV. All rights reserved.
2. Pharmacokinetic Properties
terminal elimination half-life of agomelatine is 2.3hours.
Hepatic impairment drastically increased the sys- temic exposure to agomelatine in a single-dosestudy.[31] Agomelatine is absorbed rapidly after oral ad- ministration, with the maximum plasma concentra- 3. Therapeutic Efficacy
tion being observed between 1 and 2 hours afteradministration. The absorbed fraction is >78%. In The efficacy of agomelatine in the treatment of vitro, the plasma-to-blood concentration ratio of depression has been investigated in several double- agomelatine is 1.5, showing a preferential distribu- blind, randomised, placebo-controlled studies. Pa- tion into plasma. In plasma, agomelatine is >95% tients had to fulfill DSM-IV[32] criteria for MDD (or protein bound irrespective of concentration and this bipolar II disorder [depressed] in the study by Loo et is not modified by age or in patients with renal al.[33]) to be eligible for these studies. In these stud- impairment. Under normal protein concentrations, ies, the primary endpoint evaluated was change in albumin and α-1 acid glycoprotein contributed 35% the 17-item HAM-D score from baseline. Secondary and 36%, respectively, to the binding of agomelatine endpoints included changes in the Montgomery- in whole blood. Agomelatine is moderately distrib- Asberg Depression Rating Scale (MADRS), the uted throughout the body, with a volume of distribu- Clinical Global Impression-Severity of Illness (CGI-S) score, and HAM-D depressed mood andpsychic anxiety item scores. Endpoints were as- sessed in the intention-to-treat (ITT) population us-ing the last observation carried forward (LOCF) Agomelatine is metabolised by the liver and the analysis in the observed cases (OC) population.
metabolites are excreted mainly in the urine. The The antidepressant effects of agomelatine appear drug is metabolised by 7-O-demethylation (leading to be mediated mainly through its action as an to S-21517), hydroxylation (mainly leading to antagonist at the 5-HT2C receptor subtype, although S-21540) and the formation of 3,4-dihydrodiol its agonist effect at MT1/MT2 receptors may also (S-22380). The metabolite S-21517 has about the contribute to some of its antidepressant proper- same affinity for 5-HT2C receptors as agomelatine, ties.[7,10] Furthermore, because of its agonist activi- while the S-22380 and S-21540 metabolites have no ties at melatonin receptors (MT1 and MT2), agome- affinity for human 5-HT2C receptors. The S-21540 latine is a novel antidepressant in that it has been and S-21517 metabolites have an affinity for me- shown to affect circadian rhythms, sleep architec- latonin receptors that is at least 100-fold less than ture, total sleep time and subjective sleep quality in the parent drug (dissociation constant [kd] being human and animal studies (see section 3.2).
1.14 × 10–9 and 6.56 × 10–8 mol/L, respectively).
The S-22380 metabolite has a low affinity for me- latonin receptors (kd = 7.98 × 10–7 mol/L).[29,30]These compounds represent approximately 61–81% 3.1.1 Dose-Finding Study
of the dose excreted in urine over the first 24 hours.
Loo et al.[33] performed a randomised, multicen- A smaller amount of agomelatine is excreted faecal- tre, double-blind, fixed-dose, 8-week trial in which ly after being metabolised to S-22380. The mean 711 patients (mean age 42.3 years) received agome- 2006 Adis Data Information BV. All rights reserved.
latine 1, 5 or 25 mg/day, placebo or paroxetine 20 in mean HAM-D total scores compared with place-bo.[33] mg/day. Patients had to fulfill diagnostic criteria forMDD or bipolar II disorder (depressed) according to Agomelatine at a dosage of 25 mg/day also ap- DSM-IV, and have an initial HAM-D score of 22 to peared to be effective in treating anxiety associated be enrolled. Of the patients enrolled, 67.1% met with depression, as demonstrated by reductions in DSM-IV criteria for a recurrent major depressive the Hamilton Anxiety Rating Scale. Patients treatedwith agomelatine 25 mg/day and those treated with episode and 33.5% had an episode of severe intensi- paroxetine achieved significant reductions in anxie- ty, as defined by a HAM-D score of ≥25 at inclusion ty compared with placebo, (p < 0.05, p = 0.004).[33] and by psychiatrist evaluation. This study deter- The groups treated with lower dosages of agomela- mined the optimal dosage of agomelatine for the tine (1 and 5 mg/day) did not have significant de- treatment of depression and used the established creases in anxiety compared with placebo.[33] antidepressant paroxetine as a study validator.
In summary, the study by Loo et al.[33] deter- Pivotal analysis demonstrated agomelatine 25 mined that agomelatine 25 mg/day was the optimal mg/day to be the optimal dosage for the treatment of antidepressant dosage tested. At this dosage, de- MDD.[33] Although there was a statistically signifi- pressed patients taking agomelatine were signifi- cant difference in the mean HAM-D final score cantly more likely to achieve remission or have a between all three dosages of agomelatine and place- good clinical response than those receiving placebo, bo (p = 0.037), subsequent complementary analyses showed an antidepressant response within 2 weeks showed that only agomelatine 25 mg/day was clear- of initiating treatment (and more quickly than those ly more effective than placebo. Compared with the receiving paroxetine), and had a reduction in anxiety paroxetine and placebo groups, there was a higher associated with depression to a similar degree as number of responders in the agomelatine 25 mg/day group (56.3% vs 46.3% vs 61.5%, respectively; 3.1.2 Additional Therapeutic Studies
responders defined as those patients having a HAM- Kennedy and Emsley[34] evaluated the effect of D improvement of ≥50% compared with baseline).
agomelatine compared with placebo in treating de- Both the paroxetine and agomelatine 25 mg/day pression in a randomised, double-blind, multicentre, groups achieved significantly higher rates of remis- flexible-dose, 6-week trial involving 212 outpatients sion than the placebo group (25.7% [p < 0.05], (mean age 42.5 years). Subjects enrolled in the study 30.4% [p < 0.01] and 15.4%, respectively; remission had MDD of moderate to severe intensity (HAM-D was defined as a HAM-D at study end of <7), with score of >22) and were initially treated with either the highest percentage of patients achieving remis- agomelatine 25 mg/day or placebo. If the depression sion being in the agomelatine 25 mg/day group.[33] failed to improve sufficiently after 2 weeks, the This study also demonstrated that agomelatine 25 dosage of agomelatine or placebo was doubled (50 mg/day is clinically effective shortly after initiating treatment. After 2 weeks of treatment, patients re- In this study, patients treated with agomelatine ceiving this dosage of agomelatine had a significant were significantly more likely to improve than those decrease in mean HAM-D total scores compared receiving placebo. Patients who received agomela- with patients receiving placebo. In contrast, patients tine (25 and 50 mg/day) had a significantly lower required 4 weeks of treatment with paroxetine mean HAM-D score at the study endpoint compared before achieving a statistically significant decrease with those who received placebo (14.1 ± 7.7 vs 16.5 2006 Adis Data Information BV. All rights reserved.
± 7.4; p = 0.026). Additionally, the percentage of sion. Additionally, increasing the dosage of agome- responders at week 6 (patients who showed a HAM- latine to 50 mg/day appeared to be effective and well D score decrease of ≥50%) was significantly higher tolerated in patients whose symptoms failed to show in the agomelatine group compared with the placebo improvement after 2 weeks on a dosage of 25 mg/ group (49.1% vs 34.3%, respectively; p = 0.03).[34] Severely depressed patients (defined as having a In another study published by Kennedy[36] the baseline HAM-D score of ≥25) had a significantly efficacy and adverse effect profile of agomelatine higher percentage of responding following treat- versus venlafaxine in the treatment of depression ment with agomelatine than placebo (48.7% vs was evaluated. This was a 12-week, multicentre, 30.7%, respectively; p = 0.024) at 6 weeks.[34] The double-blind, randomised study, which involved rate of response to agomelatine in this study was 277 patients (age range 18–60 years). Patients had to comparable to that reported in a meta-analysis of 32 meet the DSM-IV criteria for MDD (MADRS score ≥20) and were given either agomelatine 50 mg/day In patients who did not respond to agomelatine for the duration of the study, or a sustained release 25 mg/day (37%), increasing the dosage to 50 mg/ formulation of venlafaxine at a dosage of 75 mg/day day did significantly improve their depression. This for 2 weeks and then 150 mg/day for the remaining was demonstrated by significant improvements in the HAM-D score at week 6 in the agomelatine With regards to clinical efficacy, this study 25–50 mg/day patient group (from 26.1 ± 2.6 at demonstrated that patients with MDD who under- baseline to 17.5 ± 7.4) compared with the ‘increased went 12 weeks of treatment with agomelatine 50 placebo’ group (26.7 ± 2.8 to 20.4 ± 6.0; mg/day or venlafaxine 150 mg/day had similar rates p = 0.045).[34] With regard to achieving a full remis- of response (82.5% and 79.9%, respectively; re- sion (as defined by HAM-D total score <6), at the sponders defined as those achieving a 50% reduc- conclusion of the study patients treated with agome- tion from baseline MADRS score). Additionally, latine were more likely to reach this endpoint than patients in each group achieved similar rates of those receiving placebo (20.8% vs 13.3%); howev- stable remission at week 12 (78 of 137 [57%] vs 83 er, this was not a statistically significant difference of 140 [59%] subjects, respectively; stable remitters were defined as those patients who were responders Several secondary measures of depression also [50% reduction from baseline MADRS score] at demonstrated that agomelatine alleviated depression week 10 and who had a MADRS score of ≤12 at more effectively than placebo. The CGI-S scores improved significantly in patients treated with Guilleminault[37] reported on the effects of agomelatine (4.8 ± 0.7 to 3.2 ± 1.3) compared with agomelatine versus venlafaxine on depression and placebo (4.8 ± 0.07 to 3.6 ± 1.3) at week 6 subjective sleep (onset and quality) in patients treat- (p = 0.017). Although the CGI-Improvement scores ed as outpatients for MDD (n = 332). In this 6-week, tended to improve in patients treated with agomela- double-blind, randomised, multicentre study, sub- tine compared with those treated with placebo, this jects received agomelatine 25 mg/day or venlafax- did not reach statistical significance.[34] ine 75 mg/day for 2 weeks; the dosages were in- In summary, Kennedy and Emsley[34] demon- creased to 50 mg/day or 150 mg/day, respectively, strated that treatment with agomelatine 25 mg/day after week 2 for the duration of the study if symp- was more effective than placebo in treating depres- toms had not resolved.[37] This study showed that 2006 Adis Data Information BV. All rights reserved.
treatment of depressed patients with either agomela- study evaluating the effect of agomelatine on the tine 50 mg/day or venlafaxine 150 mg/day resulted circadian rhythm and sleep parameters of healthy, in comparable antidepressant efficacy after 6 weeks nondepressed older men (n = 8, mean age 60 years).
of treatment (final HAM-D score 9.9 ± 6.6 and 11.0 This study demonstrated that evening administration of agomelatine affects the circadian rhythm. In this In summary, these studies show that agomelatine study, patients received daily treatment for 15 days is an effective antidepressant, with similar response with either agomelatine 50mg or placebo. The medi- and remission rates to several other antidepressants.
cation was administered at 1830 hours and parame- Increasing the dosage of agomelatine from 25 mg/ ters assessing circadian rhythm were analysed over a day to 50 mg/day is effective in treating patients 24-hour period. Following evening administration with refractory depression several weeks after initi- of agomelatine, a phase advancement of nearly 2 hours was observed for body temperature and phaseadvances of 1.5–2.0 hours were noted for cortisol secretion. This study clearly demonstrated that eve-ning administration of agomelatine advances the Agomelatine is a unique antidepressant in that it circadian rhythm as measured with several physio- has been shown to affect sleep differently from any logical variables. Sleep parameters measured by other antidepressant. This is thought to occur be- polysomnography, including total sleep time and cause of its agonist interaction with the MT1 and sleep stages, were not significantly affected in this MT2 receptors. Five published studies have re- viewed the effects of agomelatine on human sleep Additionally, Leproult et al.[39] showed that pa- tients treated with agomelatine had increased In the already mentioned study of Guillem- growth hormone secretion during the wake period, inault[37] comparing agomelatine to venlafaxine (see with levels changing from 155 ± 41μg while patients section 3.1.2), patients treated with agomelatine re- were receiving placebo to 295 ± 75μg while receiv- ported significant improvements in ‘getting to sleep’ ing agomelatine. The clinical effects of the increase (p = 0.007) and ‘quality of sleep’ (p = 0.015) as in growth hormone level are unknown, but theoreti- assessed by the Leeds Sleep Evaluation Question- cally the increase may have beneficial effects on naire (LSEQ) compared with patients taking metabolic variables dependent on the growth hor- venlafaxine. These changes were noted 1 week after mone axis, such as muscle strength and bone metab- treatment was initiated and remained statistically significant for the duration of the study.
Cajochen et al.[38] evaluated the effects of agome- Quera-Salva et al.[40] evaluated the effects of latine on sleep architecture in a small, crossover agomelatine on sleep architecture and other sleep study utilising polysomnography. In this single-dose parameters in patients with MDD. This study in- study, eight healthy nondepressed men (mean age volved 15 outpatients (mean age 36 years) who were 23–32 years) received agomelatine 5 or 100mg treated with agomelatine 25mg at bedtime. Sleep before bedtime, and reported significantly increased architecture and other sleep parameters were mea- rapid eye movement (REM) sleep with no effect on sured by polysomnography, while subjective sleep perception was evaluated by the LSEQ. After 42 Leproult et al.[39] published a randomised, dou- days of treatment with agomelatine, sleep efficiency ble-blind, placebo-controlled, two-period, crossover increased by 4% (95% CI 0.03, 8.69), and wake after 2006 Adis Data Information BV. All rights reserved.
sleep onset (WASO) decreased from 42 to 19 min- These changes are of interest for several reasons.
utes. Slow-wave sleep increased by 16 minutes First, they indicate that agomelatine has an action on (95% CI 1.79, 26.06) and no changes were noted in NREM sleep that occurs soon after starting the REM sleep. Subjectively patients felt their sleep had medication. There is a change of sleep structure, improved with respect to ‘sleep quality’ and ‘easi- with a more consolidated sleep seen within 7 days of ness falling asleep’. These changes were recorded 7 starting the medication, as shown by the CAP days after initiation of agomelatine and lasted for the study.[41] The changes seen on polysomnography correlate with the subjective improvement of sleepand improvement of LSEQ scores. The use of two Lopes and colleagues[41] reported on the effect of visual analogue scales that assessed daytme alert- 42 days of treatment with agomelatine 25 mg/day on ness in the study of venlafaxine versus agomelatine the cyclic alternating pattern (CAP) in non-REM indicated less daytime sleepiness in patients tested (NREM) sleep (measured by polysomnography) in with agomelatine compared with patients treated 15 depressed patients. CAP was measured at base- with venlafaxine at all post-baseline visits. Interest- line and compared with CAP measured in 15 ingly, the disappearance of the NREM sleep disrup- matched controls, and also compared to results ob- tion precedes improvements in subjective mood (as tained at night 7 and night 42 of agomelatine treat- ment. This study revealed a significant decrease in score), suggesting that a component of the antide- CAP time and CAP cycle after 7 and 42 nights of pressant effect of agomelatine is mediated through treatment with agomelatine compared with the base- its ability to improve sleep structure. Although the line night. A trend toward a further decrease of CAP subjective results of sleep improvement were noted rate was seen between night 7 and night 42. At day in several larger double-blind comparative investi- 42, comparison of CAP rate, CAP time and distribu- gation trials, the results of the studies that evaluated tion of phases A between controls and subjects with sleep architecture must be viewed cautiously as the MDD were not significantly different, as opposed to polysomnographic study reported above was an what was present at baseline. These findings suggest open study and involved a small group (n = 15) of that agomelatine normalises NREM sleep in de- patients, and the CAP evaluation was performed blind and evaluated only 15 patients.
In summary, agomelatine had a significant im- pact on the sleep of patients with MDD, and this 4. Tolerability
effect was seen very early after administration of thedrug. As early as 7 days after administration, the Agomelatine was generally well tolerated, with a changes in CAP (indicated by the change in CAP good safety profile in clinical trials. Adverse effects rate and by the change in the percentage of phase A in patients receiving agomelatine occurred at a simi- subtypes of CAP, factors indicative of sleep disrup- lar rate to those in patients receiving placebo.
tion) returned to the values observed in control In the dose-finding study by Loo et al.,[33] the subjects. This suggests normalisation of NREM frequency of adverse effects such as anxiety, head- sleep.[41] Sleep efficiency was significantly im- ache, abdominal pain and diarrhoea in patients treat- proved by day 14, and all studied variables were ed with agomelatine was statistically similar to that significantly improved by day 42. The changes in in patients receiving placebo. In the reported ad- NREM sleep variables preceded the improvements verse effects there were no significant increases for seen on the 17-item HAM-D obtained in the study.
a given adverse effect in patients treated with 2006 Adis Data Information BV. All rights reserved.
agomelatine compared with those treated with pla- of serotonergic syndrome in patients taking agome- cebo. Patients who reported at least one emergent adverse event were slightly less numerous in theagomelatine 25 mg/day (51%) and placebo (54.7%) groups than in the paroxetine (66.0%) group, indi- One study evaluated discontinuation symptoms cating that agomelatine may be better tolerated than when patients abruptly stopped taking agomela- paroxetine. Additionally, the discontinuation rate in tine.[42] This was a double-blind, placebo-controlled, the agomelatine 25 mg/day group (8.0%) was simi- multicentre study that involved 192 patients receiv- lar to that in the placebo group (6.5%).[33] ing either agomelatine 25 mg/day or paroxetine 20 In the study by Kennedy and Emsley,[34] dizzi- mg/day (mean age 42.55 years). Patients received ness, nasopharyngitis and influenza were more com- either active medication or placebo during the dis- monly reported in patients treated with agomelatine continuation portion of the study. Discontinuation than placebo (9.3% vs 4.8%, 6.5% vs 3.8%, and symptoms were measured 1 and 2 weeks after ab- 6.5% vs 2.9%, respectively), although none of these ruptly stopping 12 weeks of treatment. Of the 88 differences was statistically significant. However, patients in the agomelatine group, 27 were ran- headache, nausea, fatigue, dry mouth and diarrhoea domised to have agomelatine discontinued. Forty- occurred more frequently in the placebo group. Se- three of the 104 patients treated with paroxetine vere treatment-related adverse events were reported were randomised for medication discontinuation.
by two patients in the agomelatine group (one case Patients were eligible for the study if they had of dizziness and one of pruritus) and by six patients initially fulfilled the DSM-IV diagnostic criteria for MDD and had mild to moderate depression as deter-mined by an entry MADRS score of 18–27. Further- Another study by Kennedy[36] evaluated the sexu- more, patients must have achieved sustained remis- al side effects of agomelatine utilising the Sex Ef- sion during the 12-week treatment period. Discon- fects Scale in 111 sexually active patients treated for tinuation symptoms were assessed by the total depression. Sexual dysfunction with respect to a number of discontinuation emergent signs and desire-arousal factor occurred in a significantly low- symptoms (DESS) occurring in the first and second er percentage of patients treated with agomelatine week after discontinuing the antidepressant.
than treated with venlafaxine 150 mg/day (20% vs No statistically significant difference in the num- 41.2%, respectively; p = 0.015).[36] In addition, a ber of emergent discontinuation symptoms was seen significantly lower number of patients treated with 1 week after treatment interruption between patients agomelatine complained of orgasm dysfunction as discontinuing agomelatine and those continuing compared with those treated with venlafaxine (20% agomelatine (3.0 ± 4.2 and 4.4 ± 5.7, respectively; vs 47%, respectively; p < 0.002). However, this p = 0.250). In contrast, patients discontinuing parox- study did not report if the sexual dysfunction report- etine experienced significantly more symptoms than ed in the agomelatine or venlafaxine groups re- those continuing paroxetine (7.3 ± 7.1 and 3.5 ± 4.1, flected emergent or residual symptoms.
respectively; p < 0.001).[21] Two weeks after treat- In summary, agomelatine appears to be well tol- ment interruption, there were no statistically signifi- erated in patients with depression, with an ad- cant differences in the number of observed emer- verse effect profile similar to that of placebo. In gent discontinuation symptoms between patients all treatment studies, there were no reported cases discontinuing agomelatine and those continuing 2006 Adis Data Information BV. All rights reserved.
agomelatine (2.0 ± 2.3 and 3.0 ± 4.4, respectively; fects of these sleep parameters require further inves- p = 0.312). However, discontinuation symptoms were more prevalent in patients who stopped parox- Despite these exciting and optimistic findings, etine compared with those continuing paroxetine the current published data on agomelatine are some- (6.5 ± 6.1 and 3.3 ± 3.2, respectively; p = 0.004).[21] what limited. Specifically, published information No emergent serious adverse events occurred in on efficacy and tolerability is only available from either group during the discontinuation period.[21] short-term trials; data from a 12-month study per- Additionally, relapse rates were not significantly formed recently by the drug developer are not yet higher in the subjects discontinuing treatment of available. Also, although older subjects (up to 83 agomelatine compared with those who continued years of age) have been involved in trials, a specific taking the medication.[21] This study clearly indi- study focusing only on the elderly has not yet cates that patients who abruptly discontinue agome- been reported. Similarly, systematic investigation of latine do not experience significant discontinuation the effects of agomelatine in children and teenag- ers aged 17 years and younger has not been pub-lished.
5. Conclusion
Acknowledgements
Agomelatine appears to be an effective antide- Dr Guilleminault has served as a scientific advisor to pressant with a unique mechanism of action. It is Laboratoire Servier for sleep studies. Dr Zupancic has no reported to be well tolerated and, according to com- conflicts of interest that are directly relevant to the contents of parative trials, to have an adverse effect profile that this review. No sources of funding were received for the is related to traditional and some newer antidepres- sants, specifically venlafaxine and paroxetine, al-though adverse effects appear to be less frequent References
1. Kessler RC, Berglund P, Demler O, et al. 2005 lifetime preva- than with these other agents. Additionally, agomela- lence and age-of-onset distributions of DSM-IV disorders in tine has positive effects on subjective reports of the National Comorbidity Survey Replication. Arch Gen Psy- sleep quality and shortens sleep latency. Agomela- 2. Avery DH, Wildschiodtz G, Rafaelsen OJ. Nocturnal tempera- tine has been shown to affect sleep architecture, as ture in affective disorder. J Affect Disorder 1982; 4: 61-71 scored by the traditional polysomnographic staging 3. Boivin DB. Influence of sleep-wake and circadian rhythm dis- turbances in psychiatric disorders. J Psychiatry Neurosci 2000; of Rechtschaffen and Kales (as shown in the studies by Cajochen et al.[38] and Quera-Salva et al.[40]) and 4. Srinivasan V, Smits M, Spence W, et al. Melatonin in mood also improves the stability of NREM sleep as evalu- disorders. World J Biol Psychiatry 2006; 7: 138-51 5. Fava M. Daytime sleepiness and insomnia as correlates of ated with usage of the more contemporary method depression. J Clin Psychiatry 2004; 65 Suppl. 16: 27-32 of CAP scoring. The changes in sleep patterns seem 6. Bonnefond C, Martinet L, Lesieur D, et al. Characterization of S-2098, a new melatonic analogue. In: Touitou Y, Arendt J, to occur early after administration of the drug and Pavet P, editors. Melatonin and the pineal gland. Amsterdam: precede the changes in HAM-D score. This is of clinical and theoretical interest, as there has been a 7. Millan MJ, Gobert A, Lejeune F, et al. The novel melatonin agonist agomelatine (S20098) is an antagonist at 5-hydroxy- suggestion that chronic sleep disruption and chronic tryptamine 2C receptors, blockade of which enhances the insomnia can lead to depression; the demonstration activity of frontocortical dopaminergic and adrenergic path-ways. J Pharmacol Exp Ther 2003 Sep; 306 (3): 954-64 that improvement of sleep precedes improvement of 8. Yous S, Andrieux J, Howell HE, et al. Novel naphthalenic mood would be an important fact with potential ligands with high affinity for the melatonin receptor. J Med therapeutic implications. However, the clinical ef- 2006 Adis Data Information BV. All rights reserved.
9. Cussac D, Newman-Tancredi A, Pasteau V, et al. Antagonist 26. Armstrong SM, McNulty OM, Guardiola-Lemaitre B, et al.
properties of the melatonin agonist S 20098 (agomelatine) at Successful use of S20098 and melatonin in an animal model of recombinant, human (h) serotonin (5-HT2c) receptors [ab- delayed sleep-phase syndrome (DSPS). Pharmacol Biochem stract]. Int J Neuropsychopharmacol 2002; 5 Suppl. 1: S68 10. Papp M, Gruca P, Boyer PA, et al. Effect of agomelatine in the 27. Redman JR, Francis AJ. Entrainment of rat circadian chronic mild stress model of depression in the rat. Neuropsy- rhythms by the melatonin agonist S-20098 requires intact suprachiasmatic nuclei but not the pineal. J Biol Rhythms 11. Bourin M, Mocaer E, Porsolt R. Antidepressant-like activity of S 20098 (agomelatine) in the forced swimming test in rodents: 28. Ying SW, Rusak B, Mocaer E. Chronic exposure to melatonin involvement of melatonin and serotonin receptors. J Psychiatry receptor agonists does not alter their effects on suprachiasmat- ic nucleus neurons. Eur J Pharmacol 1998 Jan 19; 342 (1): 29- 12. Barden N, Shink E, Labbe M, et al. Antidepressant action of agomelatine (S 20098) in a transgenic mouse model. Prog 29. Lins R. Study PKH-004: excretion balance metabolic profile, Neuropsychopharmacol Biol Psychiatry 2005 Jul; 29 (6): 908- May 1994 to July 1994. Courbevoie: Servier, 1994. (Data on 13. Bertaina-Anglade V, Mocaer E, Drieu La Rochelle C. Antide- 30. Dickens J. Study PKH-012: PK excretion balance metabolic pressant-like action of S20098 (agomelatine) in the learned profile, March 2001 to April 2001. Courbevoie: Servier, 1994.
helplessness test [abstract]. Int J Neuropsychopharmacol 2002; 31. Leclerc V. Study PKH-041: intrinsic factors: influence of liver 14. Millan MJ, Brocco M, Gobert A, et al. Anxiolytic properties of impairment on PK, November 2001 to August 2002.
agomelatine, an antidepressant with melatoninergic and ser- Courbevoie: Servier, 1994. (Data on file) otonergic properties: role of 5-HT2C receptor blockade.
32. American Psychiatric Association. Diagnostic and statistical Psychopharmacology (Berl) 2005; 177: 448-58 manual of mental disorders. 4th edition. Washington, DC: 15. Papp M, Litwa E, Gruca P, et al. Anxiolytic-like activity of agomelatine and melatonin in three animal models of anxiety.
33. Loo H, Hale A, D’Haenen H. Determination of the dose of agomelatine, a melatoninergic agonist and selective 5-HT(2C) 16. Bremner JD, Narayan M, Anderson ER, et al. Hippocampal antagonist, in the treatment of major depressive disorder: a volume reduction in major depression. Am J Psychiatry 2000 placebo-controlled dose range study. Int Clin Psychopharma- 17. Sheline YI. Neuroimaging studies of mood disorder effects on 34. Kennedy SH, Emsley R. Placebo-controlled trial of agomelatine the brain. Biol Psychiatry 2003 Aug 1; 54 (3): 338-52 in the treatment of major depressive disorder. Eur Neuropsy- 18. Fuchs E, Flugge G, Czeh B. Remodeling of neuronal networks by stress. Front Biosci 2006 Sep 1; 11: 2746-58 35. Storosum JG, Elferink AJ, van Zwieten BJ, et al. Short-term 19. Vythilingam M, Vermetten E, Anderson GM, et al. Hippocam- efficacy of tricyclic antidepressants revisited: a meta-analytic pal volume, memory, and cortisol status in major depressive study [published erratum appears in Eur Neuropsychopharma- disorder: effects of treatment. Biol Psychiatry 2004 Jul 15; 56 col 2001; 11: 325]. Eur Neuropsychopharmacol 2001; 11: 173- 20. Eriksson PS, Perfilieva E, Bjork-Eriksson T, et al. Neurogenesis 36. Kennedy SH. Sexual function in remitted depressed patients in the adult human hippocampus. Nat Med 1998 Nov; 4 (11): following agomelatine and venlafaxine XR treatment [ab- stract]. Eur Neuropsycopharmacol 2005; 15 Suppl. 3: S440 21. Duman RS, Monteggia LM. A neurotrophic model for stress- 37. Guilleminault C. Efficacy of agomelatine versus venlafaxine on related mood disorders. Biol Psychiatry 2006 Jun 15; 59 (12): subjective sleep of patients with major depressive disorder [abstract]. Eur Neuropsychopharmacol 2005; 15: Suppl. 3: 22. Banasr M, Soumier A, Hery M, et al. Agomelatine, a new antidepressant, induces regional changes in hippocampalneurogenesis. Biol Psychiatry 2006 Jun 1; 59 (11): 1087-96 38. Cajochen C, Krauchi K, Mori D, et al. Melatonin and S-20098 increase REM sleep and wake-up propensity without modify- 23. Redman JR, Guardiola-Lemaitre B, Brown M, et al. Dose ing NREM sleep homeostasis. Am J Physiol 1997; 272 (4 Pt dependent effects of S-20098, a melatonin agonist, on direc- tion of re-entrainment of rat circadian activity rhythms.
Psychopharmacology (Berl) 1995; 118: 385-90 39. Leproult R, Van Onderbergen A, L’hermite-Baleriaux M, et al.
24. Martinet L, Guardiola-Lemaitre B, Mocaer E. Entrainment of Phase-shifts of 24-h rhythms of hormonal release and body circadian rhythms by S-20098, a melatonin agonist, is dose and temperature following early evening administration of the plasma concentration dependent. Pharmacol Biochem Behav melatonin agonist agomelatine in healthy older men. Clin 25. Van Reeth O, Weibel L, Olivares E, et al. Melatonin or a 40. Quera-Salva MA, Vanier B, Chapotot F, et al. Effect of agome- melatonin agonist corrects age-related changes in circadian latine on the sleep EEG in patients with major depressive response to environmental stimulus. Am J Physiol Regul In- disorder [abstract]. Eur Neuropsychopharmacol 2005; 15 Sup- 2006 Adis Data Information BV. All rights reserved.
41. Lopes MC, Quera-Salva MA, Guilleminault C. Cyclic alternat- double-blind, placebo-controlled discontinuation study. Int ing pattern in the NREM sleep of patients within major depres- sion disorder: baseline results and change overtime with anew antidepressant: agomelatine. Sleep Med 2005; 6 Suppl. 2:87-8 Correspondence and offprints: Dr Christian Guilleminault, 42. Montgomery SA, Kennedy SH, Burrows GD, et al. Absence of Stanford University Sleep Medicine Program, 401 Quarry discontinuation symptoms with agomelatine and occurrence ofdiscontinuation symptoms with paroxetine: a randomized, Road, Suite 3301, Stanford, CA 94305, USA.
2006 Adis Data Information BV. All rights reserved.

Source: http://mpa.my/file_dir/1229889203468df002645ef.pdf