Evidence-Based Pharmacotherapy of Generalised Anxiety Disorder: Focus on Agomelatine

This article is published with digital features, including a video abstract, to facilitate understanding of the article. To view digital features for this article go to https://​doi.​org/​10.​6084/​m9.​figshare.​14959524.

Recent network meta-analyses (NMAs) provide useful overviews of the available evidence to support different classes of anxiolytic drug, and individual drugs within those classes [15‐17]. In an analysis of 57 clinical trials, Chen et al. found that all classes of pharmacotherapy for GAD were more efficacious (based on changes in Hamilton Anxiety Rating Scale [HAM-A]) than placebo, with the exception of serotonin-modulating drugs and SGAs [15]. The most efficacious drug classes, relative to placebo, were the norepinephrine–dopamine reuptake inhibitors, followed by the noradrenergic and specific serotonergic antidepressant. Importantly, however, the volume of data on these classes was considerably smaller than for SSRIs, SNRIs, BZDs and other classes. BZDs, as a class, showed the smallest efficacy difference versus placebo among the effective drug classes. Most pharmacological interventions had greater treatment effects than psychological interventions [15].

Separately, Slee and colleagues performed an NMA of 89 trials of individual anxiolytic agents that included more than 25,000 patients with GAD [16]. The most efficacious drugs (again, based on change in HAM-A score) were duloxetine, pregabalin, venlafaxine and escitalopram; interestingly, the SGA quetiapine was found to be more efficacious than duloxetine, but was limited by its poor tolerability. There was favourable evidence in terms of efficacy and tolerability for several other drugs, but the authors highlighted that sample sizes for these agents were small (buspirone, fluoxetine, sertraline, mirtazapine and agomelatine) [16]. Notably, this NMA did not include trials reported before 1994, and may therefore have excluded a significant volume of relevant data.

Lastly, He et al. conducted an NMA of individual first-line pharmacotherapies (SSRIs and SNRIs) for the treatment of GAD [17], and included 41 trials of at least 4 weeks’ duration that used the HAM-A as the main measure of efficacy (n > 15,000). With the exception of fluoxetine and vortioxetine, all of the drugs included in the analysis had greater anxiolytic efficacy than placebo. Consistent with the findings of Slee and colleagues [16], drugs with the greatest effect on HAM-A scores, relative to placebo, were duloxetine, escitalopram and venlafaxine [17]; pregabalin was not included in the analysis.

Other meta-analyses of pharmacotherapy for GAD have been useful in indicating that a number of classes of medication are useful for GAD [18], that anxiolytic effects persist beyond the active treatment period [19], that maintenance treatment is effective in preventing relapse [20], and that effective treatment improves quality of life [21]. Much research is needed in this area, including trials of more representative patients [22], including those with significant comorbidity [23], trials of GAD across the patients’ lifespan [24, 25], pragmatic trials [26], and trials of drugs that target novel mechanisms of action [14]. Future research may also include work based on translational neuroscience [27], epidemiological studies [28] and big data [29], and should also address the question of whether clinical characterization may improve personalized treatment [30].

The network meta-analyses discussed here emphasize that there is a robust evidence base supporting the use of established drug classes, particularly SNRIs and SSRIs, in the treatment of GAD. However, there have been few studies focused on severe GAD in particular, and there is a need for newer agents with novel mechanisms of action to ensure that there are different treatment options available for people with GAD, including those who do not respond to or cannot tolerate established therapies [31]. Among these newer agents is agomelatine, which Stein and colleagues systematically compared with placebo and active comparators in patients with GAD [31‐35]. Importantly, in the NMA by Slee and colleagues mentioned earlier, agomelatine was the only newer agent to show efficacy compared with placebo in the treatment of GAD [16].

In a preliminary 12-week study in 121 patients with GAD [35], HAM-A score curves for agomelatine (25–50 mg/day) and placebo began to separate after 4 weeks. At week 12, the estimated difference between the treatment groups (in mean change in HAM-A score vs baseline) was − 3.28 (95% confidence interval [− 6.41, − 0.15]; P = 0.04). Subsequently, agomelatine (25–50 mg/day) was found to have comparable efficacy to escitalopram (10–20 mg/day), in terms of both reductions in HAM-A scores and response rates, in a randomised placebo-controlled trial (n = 412). In this trial, escitalopram was used as the comparator in order to validate the study results. Although both active treatments were superior to placebo in the overall study group and in patients with severe GAD (defined as a baseline HAM-A score ≥ 25), fewer agomelatine than escitalopram recipients experienced severe adverse events (AEs; n = 1 vs 4, respectively) or AEs that led to treatment discontinuation (n = 3 vs 10, respectively).

Since the publication of these two studies, subsequent trials have confirmed the efficacy and safety of agomelatine in the treatment of GAD at lower daily dosages (10–25 mg/day) [33], and in a head-to-head trial versus escitalopram in severe GAD [32]. In the latter study, both treatments were associated with clinically significant decreases in HAM-A scores at week 12, and were thus efficacious in treating patients with severe GAD. There were no significant differences between agomelatine and escitalopram in terms of the size of the reductions in HAM-A score from baseline, in either the overall study group (mean ± standard deviation, − 16.0 ± 9.1 vs − 16.9 ± 8.4, respectively) or in more severely anxious patients (− 16.4 ± 9.5 vs − 17.4 ± 8.6, respectively; Fig. 1) [32]. However, the pre-established criteria for non-inferiority of agomelatine versus escitalopram were not met (HAM-A total score estimate, − 0.91; standard error, 0.69; 95% confidence interval − 2.26, 0.44; P = 0.195). Agomelatine was associated with a numerically lower incidence of AEs compared with escitalopram (46.9% vs 58.8%) [32].

Maintenance agomelatine treatment has been shown to be more effective than placebo in preventing relapse in patients with GAD [34]. Over a 6-month period, 19.5% of patients receiving agomelatine (25–50 mg/day) experienced relapse, compared with 30.5% of placebo recipients [hazard ratio (HR) 0.582; P = 0.046]. Agomelatine was also associated with a significantly lower relapse rate versus placebo in patients with severe GAD (20.8% vs 42.9%; HR 0.407; P = 0.006).

Agomelatine has also been found to be efficacious in anxious depression, as discussed in a separate article in this supplement. Symptoms of anxiety are present in 40–50% of patients with major depressive disorder (MDD), and are associated with greater functional impairment and a higher risk of suicide in this patient group [36]. In a pooled analysis of six trials in patients with MDD, agomelatine was found to have a significantly greater effect on HAM-A scores over 6–8 weeks compared with placebo (between-group difference at endpoint, 3.13 points; P < 0.005), and a comparable or greater effect versus fluoxetine (1.24 points; P = 0.137), sertraline (2.36 points; P = 0.016) or venlafaxine (0.82 points; P = 0.327) [36]. Findings were similar for the subgroup of patients with more severe anxiety symptoms.

In current guidelines for the management of GAD, SSRIs and SNRIs are typically recommended as first-line agents of choice [1‐6]. Agomelatine has also demonstrated efficacy in anxiety symptoms in MDD [36], as well as in GAD [32, 34, 37]. Considering its evidence base in GAD to be comparable to that for escitalopram, the Anxiety Disorders Association of Canada recommends agomelatine as a first-line agent [4]. Agomelatine is considered as a second-line agent (in patients with GAD resistant to SSRI or SNRI treatment) by the Royal Australian and New Zealand College of Psychiatrists [1] and the South African Society of Psychiatrists [6]. Thus, agomelatine has an established place in therapy as an alternative to initial treatment with an SSRI or SNRI, and as an option in patients who do not respond, or have only a partial response, to first-line GAD. Its use in GAD would be off-label since it is not approved for use in this indication.