Abstract
Agomelatine (Valdoxan), a synthetic melatonergic receptor agonist at the MT1 and MT2 receptors, was first used in the management of sleep disorder. Its 5HT2C receptor antagonistic properties support its antidepressant potential. It is currently licensed in the UK, Europe and USA for the treatment of major depressive disorder. Although the randomized controlled evidence base for its use is growing, there are no retrospective, naturalistic studies available. We aimed to determine the tolerability and clinical effectiveness of agomelatine in unipolar depression. We also examined whether being refractory to treatment altered clinical outcome. Forty-eight patient records were examined. Twenty-five percent were treatment refractory: Clinical Global Impression (CGI) Severity score at the start of treatment was 3.81 compared with 3.38 at the end of treatment. Fifty-four percent improved at least minimally; only 12.5% were much or very much improved. Treatment-refractory patients had a poorer outcome with higher discontinuation rates and lower CGI Improvement (p = 0.0205). Treatment-refractory patients also had a higher CGI Severity score at the end of treatment than at treatment commencement (3.92 versus 3.75), although this was not statistically significant.
Introduction
Although antidepressant pharmacotherapy remains the mainstay of treatment for major depressive disorder (MDD), there are limitations to the current treatments available. It has been more than 20 years since the introduction of selective serotonin reuptake inhibitor (SSRI) antidepressants and new pharmacotherapeutic developments in the management of MDD have been slow in development. Agomelatine (Valdoxan) is the most recently licensed antidepressant in the UK for the treatment of major depressive episodes in adults. It is a synthetic melatonergic receptor agonist at the MT1 and MT2 receptors, and has serotonin receptor antagonistic properties. The evidence base for its role in the treatment of depression is growing, with short-term double-blind randomized controlled trials (RCTs) showing agomelatine to be efficacious over placebo [Olli et al. 2007; Stahl et al. 2010], sertraline [Kasper et al. 2010], fluoxetine [Hale et al. 2010] and venlafaxine [Lemoine et al. 2007].
There is also some evidence to suggest that agomelatine can be separated from placebo as early as 1 week and that a sustained advantage over placebo is seen at up to 10 months [Kennedy, 2009]. This early improvement may partly be due to the restoration of sleep architecture – especially if patients have prominent sleep dysregulation. It is known that there is a relationship between improvement in sleep-related complaints and improvement in mood [Buysee et al. 1997] and it is also becoming increasingly recognized that recurrence of a depressive episode may be preceded by the development of or the worsening of sleep disturbance [Buysee et al. 1997; Armitage et al. 2002]. The early occurrence of the first rapid eye movement (REM) sleep period is an important change in the architecture of a depressed person’s sleep pattern [Benca et al. 1992]; however, non-REM sleep changes also occur [Buysee et al. 1997]. One study looking at the importance of non-REM sleep disturbance in MDD using cyclic-alternating pattern analysis found that agomelatine had a fast and positive effect on non-REM sleep disturbance, which was then associated with a subjective and objective [reduction in the Hamilton Depression Rating Scale (HAM-D) score] improvement in mood [Lopes et al. 2007]. There is also some evidence to suggest that agomelatine improves sleep efficiency and increases the total amount of slow wave sleep [Quera Salva et al. 2007]. It is also thought to improve ‘daytime alertness’ compared with venlafaxine [Lemoine et al. 2007]. Agomelatine has also been shown to prevent relapse at similar rates to other antidepressants [Goodwin et al. 2009]. It appears to be well tolerated, with less weight gain, sexual side effects, sleep problems and withdrawal syndrome compared with SSRIs [European Public Assessment Report; Montgomery et al. 2006; Rouillon, 2006].
There is evidence that agomelatine improves Hamilton Anxiety Rating Scale scores in patients with generalized anxiety disorder [Stein et al. 2008]. In patients with seasonal affective disorder, 25 mg of agomelatine over 14 weeks was associated with significant improvements in the Structured Interview Guide for the HAM-D, with a particular improvement in sleep disturbance and daytime fatigue [Pjrek et al. 2007]. Adjunctive use in patients with bipolar type 1 disorder also showed improvement in HAM-D scores [Calabrese et al. 2007] and the risk of triggering a manic switch was similar to that seen with venlafaxine [Calabrese et al. 2007]. This small study involved 21 patients with type I bipolar disorder who were prescribed concurrent mood-stabilizing agents (either lithium or valporic acid). All the patients were followed up for 6 weeks and although the preliminary findings suggest 25 mg of agomelatine is efficacious, a RCT is planned to confirm the results.
Given its apparent efficacy and favourable side-effect profile, some have regarded agomelatine as an ‘innovative treatment for major depressive disorder patients offering a new approach in the management of depressed patients’ [den Boer et al. 2006]. However, most of the evidence about the efficacy and tolerability of agomelatine as an antidepressant is derived from RCTs which are of short duration (usually 6–12 weeks). Currently there are no published retrospective, naturalistic studies involving agomelatine in clinical practice and there are no published studies looking at combinations of agomelatine with other antidepressants.
Aims
We aimed to determine the tolerability and clinical effectiveness of agomelatine in unipolar depression in clinical practice. We also aimed to discern whether being refractory to previous treatment altered outcome.
Method
A retrospective analysis of all psychiatric contacts in a discrete geographical area in Scotland was undertaken searching the electronic records using the keyword ‘agomelatine’. The electronic records cover all secondary care psychiatric contacts in NHS Lanarkshire, which compromises a population of 550,000. The electronic records were phased into NHS Lanarkshire’s mental health service over the period 2002–2005 (initially the Motherwell/Clydesdale district in 2002, Hairmyres/East Kilbride in 2004 and the Monklands district in 2005). General, rehabilitation, liaison, addiction and forensic psychiatry services in these areas all use the electronic record system. Given the fact that agomelatine is a relatively new medication which is unlikely to be initiated in primary care, we are confident that this method of recruitment would have a high capture rate of all patients prescribed agomelatine in Lanarkshire. Patients with a diagnosis of unipolar depression were included (F32 and F33). All other International Classification of Diseases, 10th revision diagnoses were excluded.
Treatment-refractory or treatment-resistant depression has been variously defined [Stimpson et al. 2002; Svenja et al. 2005] and currently there is no consensus for its definition. It is important to differentiate between chronic depression and depression that is truly treatment refractory or treatment resistant. Although it is commonly defined as a failure to respond to at least two trials of evidence-based antidepressant therapy at an adequate dose for an adequate duration of treatment with adequate compliance, it is generally agreed that more studies are required to further define and conceptualize this phenomenon [Berlim et al. 2007]. For the purposes of our study we defined treatment resistance in clinical terms as a history of electroconvulsive therapy (ECT) or lithium prescription. This definition was chosen because both treatment options are well supported in The Maudsley as first-choice treatments for refractory depression [Taylor et al.].
Patterns of psychotropic medication coprescription were noted. Effectiveness was measured by retrospective assignment of Clinical Global Impression (CGI) scores. This procedure has been used by others for examining clinical response to other psychotropic medications [Barbee et al. 2004; Centorrino et al. 2005; Shajahan et al. 2008]. CGI as a clinical research tool has been used for approximately 30 years. It has been shown to correlate well with other well known standard research drug efficacy scales, including the HAM-D [Busner et al. 2007]. Discontinuation of treatment and hospital admission were also used as relapse indicators. Patients defined as treatment refractory were subanalysed to determine the effect of this on the relapse markers.
Results
Forty-eight patients were included. Thirty-eight percent were men and 25% were identified as treatment refractory (either having received ECT or lithium). Patients who had received ECT treatment had also been trialled on lithium. Rates of comorbid alcohol and substance misuse in our cohort were low. No one was prescribed agomelatine under compulsory treatment measures. Average treatment duration was 10.3 months (range 0.8–42.1 months). Psychotropic medication coprescription was common, occurring in over a quarter of cases, with those identified as treatment refractory having higher rates of antipsychotic and lamotrigine coprescription.
Augmentation with a second antidepressant was not infrequent in our cohort, although this was more common in the nontreatment-refractory cohort compared with those identified as treatment refractory (33% versus 17%). Augmentation was most frequently with SSRIs (43%); however, serotonin norepinephrine reuptake inhibitors (21%), tricyclic antidepressants (14%) and noradrenergic and specific serotonergic antidepressants (7%) were also used.
Fifty-four percent of the total patient population improved at least minimally (CGI Improvement < 5). Twenty-three percent of treatment-refractory patients improved compared with 64% of those not identified as treatment refractory. Only a small minority of patients were much or very much improved (10%).
Thirty-one percent of the total patient population discontinued treatment. Discontinuation rates were higher in the treatment-refractory group (50%) compared with the nontreatment-refractory group (25%). Discontinuation due to side effects was more common in the nontreatment-refractory group (67%), while discontinuation due to inefficacy was more common in the treatment-refractory group (67%). No one in our cohort was required to discontinue agomelatine due to derangement in liver function tests.
Hospitalization rates were similar in the treatment-refractory and nontreatment-refractory cohorts at 8%.
Discussion
This is the first naturalistic retrospective chart review of agomelatine in clinical practice and provides a useful overview of its use within a discrete geographical location. It is also the first retrospective chart review specifically considering the efficacy of agomelatine for refractory cases. Agomelatine appeared to be prescribed in patients identified as treatment refractory and nontreatment refractory, with those who were treatment refractory being more likely to be prescribed a higher dose (p = 0.004) compared with those who were nontreatment refractory. Patients who were treatment refractory were less likely to discontinue agomelatine because of side effects, suggesting that it was relatively well tolerated even at higher doses. This finding is in keeping with other studies which have reported no significant increase in adverse events when comparing agomelatine 25 mg daily with agomelatine 50 mg daily [Stein et al. 2008].
We identified high rates of polypharmacy – in our cohort 62.5% (n = 30) of patients were prescribed at least one additional psychotropic medication (antidepressant, mood stabilizer, antipsychotic or anxiolytic agent). There appears to be a growing trend in psychiatric practice towards polypharmacy [Mojtabai 2010] and this phenomenon is difficult to quantify and study. Much of the evidence base behind polypharmacy is focused on antipsychotic polypharmacy and the concerns about increased mortality associated with its occurrence [Waddington et al. 1998]. There is evidence for combining antidepressants [Shelton, 2003, Licht et al. 2002] and there is also evidence for combining antidepressants and antipsychotics in certain patients. A combination of olanzapine and fluoxetine for bipolar depression was the first antipsychotic/antidepressant combination to receive US Food and Drug Administration approval for the treatment of a mood disorder [Thase, 2005]. However, there have been concerns about the overuse of antipsychotics in patients with major depression [Wheeler et al. 2003]. It is difficult to know if our rate of psychotropic polypharmacy in unipolar depression is representative of current practice because of a lack of published evidence on this issue.
Augmentation of agomelatine with another antidepressant occurred commonly in our cohort at a rate of 29% (n = 14) and interestingly occurred more frequently in patients identified as nontreatment refractory (33% versus 17%). Augmentation with an SSRI (43%) was the most common combination used, although augmentation strategies included combination with venlafaxine, mirtazapine and tricyclics (Table 1). There were no cases of adverse events leading to hospitalization in our cohort of patients and so this would suggest that combining agomelatine with other antidepressants is relatively well tolerated. To our knowledge, all the RCTs published to date have involved agomelatine monotherapy and there have been no studies specifically looking at agomelatine use in combination with other antidepressants. Clearly a cautious approach to combining agomelatine with other antidepressants should be taken until there is more robust evidence about using agomelatine to augment more conventional antidepressant therapy.
Clinical, demographics and outcome measures.
Treatment refractory defined as a history of electroconvulsive therapy (ECT) or lithium treatment.
CGI-I, Clinical Global Impression Improvement; CGI-S, Clinical Global Impression Severity; CI, confidence interval; NS, nonsignificant; SSRI, selective serotonin reuptake inhibitor; TCA, tricyclic antidepressant.
Combination with antipsychotic medication also occurred in our cohort at a rate of 29% (n = 14) and was more frequent in the treatment-refractory cohort (58%, n = 7). All antipsychotics prescribed were atypical, and the most frequently prescribed medication was quetiapine (71.4%, n = 10). Again there were no cases of adverse events leading to hospitalization in this cohort, suggesting that agomelatine is relatively well tolerated when used in combination with atypical antipsychotics. Other relatively commonly prescribed combination medications included pregabalin (13%, n = 6) and lamotrigine (8%, n = 4).
Although RCTs provide robust evidence in terms of medication efficacy and tolerability, the stringent inclusion and exclusion criteria limit their extrapolation to routine clinical practice. Patients prescribed combination therapy are often excluded in RCTs; however, polypharmacy occurs commonly in clinical practice and is worth studying. Our review is also helpful in assessing the efficacy of agomelatine in treatment-refractory cases, again a clinically important and interesting group of patients.
Our naturalistic retrospective review, although limited by its small sample size and relatively short duration of follow up, is helpful in providing additional information about agomelatine’s efficacy in day-to-day clinical practice. It provides more information about agomelatine’s role as a potential adjuvant to other antidepressants and antipsychotics and suggests that agomelatine in combination with other medications is being used routinely in clinical practice.
Footnotes
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
The authors declare no conflicts of interest in preparing this article.