This study evaluated once-daily extended-release quetiapine fumarate (quetiapine XR) as adjunctive therapy in patients with major depressive disorder (MDD) with inadequate response to ongoing antidepressant treatment. In this 8-wk (6-wk active treatment/2-wk post-treatment drug-discontinuation/follow-up), multicentre, double-blind, placebo-controlled, Phase III study, 446 patients were randomized to quetiapine XR 150 mg/d, 300 mg/d, or placebo adjunct to ongoing antidepressant treatment. The primary endpoint was the change from randomization to week 6 in Montgomery–Åsberg Depression Rating Scale (MADRS) total score. At week 6, MADRS total scores significantly improved with quetiapine XR 300 mg/d vs. placebo (−14.7 vs. −11.7, p<0.01). Quetiapine XR 300 mg/d showed significant improvements vs. placebo for: MADRS total score from week 1 onwards; MADRS response [(⩾50% total score reduction) 58.9% vs. 46.2%, p<0.05] and remission [(total score ⩽8) 42.5% vs. 24.5%, p<0.01] rates; Hamilton Depression Rating Scale (HAMD) (−13.53 vs. −10.80, p<0.01) and Clinical Global Impression – Severity of illness (CGI-S) change (−1.52 vs. −1.23, p<0.05) at week 6. For quetiapine XR 150 mg/d, improvements were not significantly different vs. placebo, except for MADRS (weeks 1 and 2) and HAMD (week 6) total scores. Withdrawal rates due to adverse events (AEs) were: quetiapine XR 150 mg/d 11.5%, 300 mg/d 19.5%, and placebo 0.7%. The most common AEs (>10%) with quetiapine XR were dry mouth, somnolence, sedation, dizziness, constipation, nausea, insomnia, headache, and fatigue. In this study, quetiapine XR 300 mg/d as adjunctive therapy in patients with MDD with an inadequate response to ongoing antidepressant treatment was effective at week 6. However, the difference from placebo for quetiapine XR 150 mg/d at week 6 was not statistically significant. Both doses studied (150 and 300 mg/d) were effective at week 1 and generally well tolerated.
Major depressive disorder (MDD) is highly prevalent, chronic and disabling (Hasin et al.2005; Prince et al.2007). Despite the large number of agents licensed for MDD, response rates with initial therapies are estimated to be around 50%, while remission rates range from 30% to 40% (Nemeroff, 2007). In addition to suboptimal efficacy outcomes, 16–22% of patients with MDD report poor tolerability with initial treatments (Masand, 2003). For the 60–70% of patients who do not experience remission, residual depressive symptoms increase the burden of MDD and may also be associated with greater morbidity, mortality, healthcare resource usage, and costs (Crown et al.2002; Trivedi, 2006).
Following partial or non-response to an adequate dose of a major antidepressant taken for a sufficient duration (∼6 wk), strategies include switching, combination with another antidepressant that has a different mechanism of action, or augmentation with a non-antidepressant drug (Bauer et al.2007; Kennedy et al.2001). In randomized, controlled studies in treatment-resistant depression (defined as a lack of response to two trials of different antidepressants at an adequate dose for a sufficient duration), lithium, bupropion, buspirone, modafinil, anticonvulsants, or triiodothyronine were effective as augmentation agents (Nemeroff, 2007).
Recent placebo-controlled studies have shown that adjunctive risperidone (Mahmoud et al.2007) and adjunctive aripiprazole (Berman et al.2007, 2009; Marcus et al.2008) provide additional benefits to antidepressant monotherapy in patients with treatment-resistant MDD. Additionally, olanzapine in combination with fluoxetine (OFC) is more effective than either agent alone (Thase et al.2007). The US Food and Drug Administration (FDA) has approved aripiprazole adjunct to antidepressants for MDD and OFC for treatment-resistant MDD.
Quetiapine has demonstrated efficacy in treating depressive symptoms in schizophrenia (Buckley, 2004), bipolar depression (Calabrese et al.2005; Thase et al.2006) and as an adjunct to antidepressants for MDD with comorbid anxiety (McIntyre et al.2007; Yargic et al.2004). In addition, quetiapine is FDA-approved for schizophrenia, bipolar mania, bipolar depression, and bipolar maintenance. More recently, extended-release quetiapine fumarate (quetiapine XR) has shown efficacy as an adjunct to existing antidepressant therapy (Bauer et al.2009) and as monotherapy in patients with MDD (Cutler et al.2009; Datto et al.2008; El-Khalili et al.2009; Katila et al.2008; Weisler et al.2009), with improvement across a range of depressive symptoms.
To date, adjunct aripiprazole and OFC have been investigated in patients with treatment-resistant depression and these study designs included a prospective treatment failure (Berman et al.2007, 2009; Marcus et al.2008; Thase et al.2007). However, for patients with partial or non-response in clinical practice, guidelines recommend optimizing the current treatment regimen (dose increase, switching/augmenting) (Bauer et al.2007; Kennedy et al.2001). These recommendations provide the rationale for the patient population enrolled in the present study [D1448C00006 (Pearl)], which was designed to evaluate once-daily quetiapine XR as adjunct treatment in patients with MDD and a history of inadequate response to ongoing antidepressant therapy.
This 8-wk, multicentre, double-blind, randomized, parallel-group, placebo-controlled Phase III study [D1448C00006 (Pearl); NCT00326105] comprised an enrolment/washout period of ⩽14 d (for the discontinuation of prohibited medications), a 6-wk randomized treatment period, and a 2-wk drug-discontinuation/follow-up period.
In accordance with the Declaration of Helsinki, International Conference on Harmonization/Good Clinical Practice and applicable regulatory requirements, Institutional Review Board or Independent Ethics Committee approval was obtained at each study centre. Following randomization, study visits occurred at weeks 1, 2, 4, and 6.
Male or female outpatients (aged 18–65 yr) with a DSM-IV diagnosis of MDD (single episode 296.2x or recurrent 296.3x; confirmed by the Mini-International Neuropsychiatric Interview), Hamilton Depression Rating Scale (HAMD) total score ⩾20, and item 1 (depressed mood) score ⩾2 at enrolment and randomization were eligible for inclusion. Patients were required to have an inadequate response during their current depressive episode to one of the following antidepressants: amitriptyline, bupropion, citalopram, duloxetine, escitalopram, fluoxetine, paroxetine, sertraline, or venlafaxine. An inadequate response was defined as continuing depressive symptoms following ⩾6 wk of therapy at adequate doses (minimum effective dose according to US label and including ⩾1 dose increase as permitted by label).
Patients were excluded from the study if they had a: DSM-IV Axis I diagnosis other than MDD within 6 months prior to enrolment; DSM-IV Axis II diagnosis that significantly impacted their current psychiatric status; current MDD episode lasting >12 months or occurring <4 wk from enrolment; history of substance or alcohol abuse or dependence as defined by DSM-IV criteria within 6 months prior to enrolment; clinically significant medical illness; risk of suicide or homicide; HAMD item 3 score ⩾3; or a suicide attempt within the past 6 months. Patients requiring psychotherapy (other than supportive psychotherapy) during the study were also excluded, unless it had been ongoing for ⩾3 months before randomization.
Patients were randomized (in a 1:1:1 ratio) in a non-centre-specific manner using a computer-based system to receive 6-wk double-blind, double-dummy treatment with one of the following three treatment regimens as adjunct to their ongoing antidepressant treatment: quetiapine XR 150 mg/d (3×50 mg tablets), quetiapine XR 300 mg/d (1×300 mg tablet), or placebo. Quetiapine XR 50 mg and 300 mg tablets were identical in appearance, smell, and taste to their respective placebo tablets. Study treatments were administered orally, once daily in the evening. Treatment adherence was assessed at each study visit based on the returned tablet count.
Titration of quetiapine XR to target dose was 50 mg/d on days 1–2, 150 mg/d on days 3–4, and 300 mg/d on day 5. Ongoing antidepressant treatment was maintained at the same dose throughout the study.
Use of other psychoactive medication was not allowed, with the exception of ongoing hypnotics to treat insomnia. Sleep medication, including benzodiazepines (⩽2 mg/d lorazepam equivalent), could be continued if it had been used for ⩾28 d before enrolment. Any anticholinergic medication was permitted for the treatment of emerging extrapyramidal symptoms (EPS), but not prophylactically.
The primary endpoint was change from randomization to week 6 in Montgomery–Åsberg Depression Rating Scale (MADRS) total score (Montgomery & Åsberg, 1979).
Secondary endpoints included change from randomization at each assessment in MADRS total scores; change from randomization at weeks 1 and 6 in MADRS individual item scores; MADRS response (⩾50% reduction in total score from randomization), and remission (total score ⩽8) rates at week 6. Post-hoc analysis of MADRS remission rates was also performed using MADRS total scores ⩽10 and ⩽12 at week 6.
The change from randomization at week 6 in HAMD total score, HAMD item 1 (depressed mood) score, Hamilton Anxiety Rating Scale (HAMA) total score (Hamilton, 1959), Clinical Global Impression – Severity of illness (CGI-S) [National Institutes of Mental Health (NIMH), 1970] score, and the proportion of patients with CGI – Improvement (CGI-I) (NIMH, 1970) score of 1 (‘very much improved’) or 2 (‘much improved’) were also assessed.
Investigators and study personnel received certified and standardized training to ensure consistency throughout the study. To minimize scoring variability, it was recommended that the same trained rater conducted all assessments for a given patient on a specific scale. The κ values (used to verify rater reliability) for MADRS assessments at baseline and at follow-up were 0.842 and 0.739, respectively.
Additional secondary endpoints were change from randomization at week 6 in: Quality of Life Enjoyment and Satisfaction Questionnaire – Short Form (Q-LES-Q-SF; Endicott et al.1993) percent maximum total score, Q-LES-Q-SF satisfaction with medication (item 15) score and Q-LES-Q-SF overall quality of life (item 16) score, and Pittsburgh Sleep Quality Index (PSQI) global score.
Safety and tolerability
Adverse events (AEs) and withdrawals due to AEs were recorded throughout the study. Barnes Akathisia Rating Scale (BARS; Barnes, 1989) and Simpson–Angus Scale (SAS; Simpson & Angus, 1970) scores were evaluated at randomization and at week 6. The incidence of suicidality was measured using a system similar to that established by Columbia University (Posner et al.2007), which searched for strings of letters within Medical Dictionary for Regulatory Activities (MedDRA) preferred AE terms that may be associated with suicidality, and by assessment of MADRS item 10 (suicidal thoughts) scores.
During the 2-wk drug-discontinuation/follow-up phase, discontinuation symptoms were assessed using a modified 18-item Treatment Discontinuation Signs and Symptoms (TDSS) scale (Michelson et al.2000), which included the additional AEs vomiting, nausea, and insomnia. Baseline TDSS scores were collected at the final randomized treatment period visit (day 43). TDSS assessments were made on post-treatment days 1, 3, 5, 7, and 14. TDSS total scores were calculated by summing the number of patient-reported treatment-emergent symptoms (TDSS items) present.
Measurements of vital signs, including weight, and fasting glucose, fasting lipid, and other laboratory parameters were obtained at each visit. Twelve-lead electrocardiogram (ECG), clinical chemistry, and haematology assessments were performed at screening and at week 6.
The modified intent-to-treat (MITT) population (randomized patients who received study drug, and had randomization and ⩾1 post-randomization MADRS total scores) was used for analysis of primary and secondary variables. For all efficacy analyses, missing data were handled using a last observation carried forward (LOCF) approach.
The target sample size was based on an expected difference in the change in MADRS total score from randomization to week 6 between quetiapine XR and placebo of 3.5 points and a standard deviation (s.d.) of 9 points. For 90% power, 140 evaluable patients per group would be required (two-sided test at a 5% significance level, i.e. α=0.05). Based on previous studies, it was expected that 93% of patients assigned to randomized treatment would be evaluable; therefore, approximately 450 randomized patients were required.
The primary analysis of MADRS total score change from randomization at week 6 was analysed using an analysis of covariance (ANCOVA) model that included centre as a random effect, treatment group as a fixed effect, and baseline MADRS total score as a covariate. The per protocol (PP) population was also examined as a robustness analysis for the primary analysis. In addition, a mixed-model repeated-measures (MMRM) analysis was performed on the change from randomization in the MADRS total score to assess the robustness of the LOCF approach for the MITT analysis. Because two dose groups were compared with placebo, the primary analysis was adjusted for multiplicity using the Simes–Hommel procedure (Hommel, 1988). The change in Q-LES-Q-SF percent maximum total score from randomization at week 6 was also adjusted for multiplicity.
Overall, 659 patients were screened at 56 centres; 446 patients were randomized to receive quetiapine XR 150 mg/d (n=148), quetiapine XR 300 mg/d (n=150), or placebo (n=148) at 52 centres in the USA between 19 April 2006 and 18 July 2007. Figure 1 illustrates the disposition of patients during the study. Of the randomly assigned patients, 445 received treatment (safety population) and 432 patients were analysed for efficacy (MITT population) after 13 patients were excluded due to missing/invalid randomization or post-randomization MADRS scores.
The MITT treatment groups were well matched in terms of demographic and clinical characteristics at randomization and the antidepressants used as adjunct therapy (Table 1). In addition, all three groups had comparable proportions of patients receiving the different classes of ongoing antidepressant [selective serotonin reuptake inhibitors (SSRIs), serotonin norepinephrine reuptake inhibitors (SNRIs), or other]. The majority of patients had a DSM-IV diagnosis of recurrent MDD (92.6%). The proportion of patients completing the 6-wk study and reasons for early withdrawal are given in Figure 1.
MDD, Major depressive disorder; MADRS, Montgomery–Åsberg Depression Rating Scale; HAMD, Hamilton Depression Rating Scale; HAMA, Hamilton Anxiety Rating Scale; CGI-S, Clinical Global Impression – Severity of illness; Q-LES-Q-SF, Quality of Life Enjoyment and Satisfaction Questionnaire – Short Form; PSQI, Pittsburgh Sleep Quality Index.
One patient in the quetiapine XR 300 mg/d treatment group received both bupropion and sertraline.
Adherence to study medication was consistent across treatment groups: 96.5% (quetiapine XR 150 mg/d), 97.3% (quetiapine XR 300 mg/d), and 97.2% (placebo) of patients in the MITT population were classified as adherent; similar results were obtained for the safety population.
Mean doses of antidepressants at randomization and throughout the study, and mean duration of treatment with each antidepressant, are shown in Table 2. Before study entry, 54.4%, 27.2%, 19.3%, and 1.1% of patients were receiving a SSRI, a SNRI, bupropion, and amitriptyline, respectively. No patients were recorded as requiring anticholinergic medication for treatment-emergent EPS during the course of the study; sleep medication was used by ⩽11.3% of patients in the quetiapine XR 150 mg/d group, ⩽6.8% in the quetiapine XR 300 mg/d group, and ⩽5.4% in the placebo group.
Duration of antidepressant was adjusted to 6 months prior to the start of the study when no start date was provided.
Mean change in MADRS total score from randomization at week 6 (primary endpoint) was significantly greater with quetiapine XR 300 mg/d than with placebo (−14.70 vs. −11.70, p<0.01). Mean MADRS total score was also reduced with quetiapine XR 150 mg/d at week 6, but this difference was not statistically significant compared to placebo (−13.60, p=0.067) (Fig. 2). Following adjustment for multiplicity, the level of significance was p=0.067 for quetiapine XR 150 mg/d and p=0.008 for quetiapine XR 300 mg/d. At week 1, mean changes in MADRS total score from randomization were −9.06 (quetiapine XR 150 mg/d, p<0.001), −8.20 (quetiapine XR 300 mg/d, p<0.01) vs. −5.95 (placebo) (Fig. 2).
The PP population analysis of the primary outcome variable confirmed the efficacy of the 300 mg/d dose from the MITT population analysis and also demonstrated a significant reduction in MADRS total scores at week 6 with quetiapine XR 150 mg/d. For the PP population, the mean change in MADRS total score from randomization at week 6 was −14.29 for quetiapine XR 150 mg/d (p<0.05), −15.44 for quetiapine XR 300 mg/d (p<0.01) vs. −11.89 for placebo; a similar pattern was observed using MMRM analysis of observed case data [−14.28 (p<0.01), −15.95 (p<0.001) vs. −11.72, respectively).
At week 1, changes that were significantly greater than placebo occurred for the following MADRS items: reduced sleep and pessimistic thoughts (both doses) and concentration difficulties (quetiapine XR 150 mg/d) (Fig. 3a). At week 6, significantly greater changes with quetiapine XR 300 mg/d than placebo occurred for the following MADRS items: inner tension, reduced sleep, pessimistic thoughts, and suicidal thoughts (Fig. 3b). For quetiapine XR 150 mg/d, a significant improvement in item 4 (reduced sleep) compared to placebo was observed at week 6 (Fig. 3b).
At week 1, MADRS response rates were 27.0% (quetiapine XR 150 mg/d, p<0.05) and 26.6% (quetiapine XR 300 mg/d, p<0.05) vs. 14.6% (placebo) (Fig. 4a). MADRS response rates at week 6 were 51.7% (quetiapine XR 150 mg/d, p=0.329) and 58.9% (quetiapine XR 300 mg/d, p<0.05) vs. 46.2% (placebo) (Fig. 4a). Using these MADRS response data, NNT values for quetiapine XR 150 mg/d and 300 mg/d were 18 and 8, respectively.
At week 6, MADRS remission rates (total score ⩽8) were 35.0% (quetiapine XR 150 mg/d, p=0.059) and 42.5% (quetiapine XR 300 mg/d, p<0.01) vs. 24.5% (placebo) (Fig. 4b). When a remission criterion of MADRS total score ⩽10 was applied post-hoc, remission rates at week 6 were: 42.0% (quetiapine XR 150 mg/d, p=0.132), 52.7% (quetiapine XR 300 mg/d, p<0.001) vs. 32.9% (placebo). When remission was defined as MADRS total score ⩽12, rates were 49.0% (quetiapine XR 150 mg/d, p=0.279), 55.5% (quetiapine XR 300 mg/d, p<0.05) vs. 42.0% (placebo).
The results for other secondary efficacy variables are shown in Table 3. For quetiapine XR 300 mg/d, improvements at week 6 vs. placebo were observed in HAMD total, HAMA total, CGI-S total, and PSQI global scores, and in the proportion of patients with a CGI-I score ⩽2. For quetiapine XR 150 mg/d, improvements in these variables were not found to be different from placebo at week 6, with the exception of HAMD total and PSQI global scores.
LSM, Least squares mean; AD, antidepressant; HAMD, Hamilton Depression Rating Scale; HAMA, Hamilton Anxiety Rating Scale; CGI-S, Clinical Global Impression – Severity of illness; CGI-I, Clinical Global Impression – Improvement; Q-LES-Q-SF, Quality of Life Enjoyment and Satisfaction Questionnaire – Short Form; PSQI, Pittsburgh Sleep Quality Index.
p values are vs. change from randomization at week 6 for placebo.
Safety and tolerability
The percentage of patients who withdrew from the study due to an AE was 0.7%, 11.5%, and 19.5% in the placebo, quetiapine XR 150 mg/d, and 300 mg/d groups, respectively. In addition, one patient in the placebo group withdrew prior to randomization due to ‘ECG PR prolongation’ and ‘ECG QT corrected interval prolonged’. In the quetiapine XR 150 mg/d and 300 mg/d groups, the most common AEs leading to discontinuation were sedation and somnolence: 2.0% and 3.4% in the quetiapine XR 150 mg/d group, and 7.4% and 4.0% in the quetiapine XR 300 mg/d group, respectively; no patients withdrew from the placebo group due to sedation or somnolence. Three (2.0%) patients in the quetiapine XR 300 mg/d group withdrew from the study due to akathisia.
Two serious AEs were reported during the study period. One patient experienced a transient ischaemic attack (placebo) and one patient experienced worsening of cervical spondylitis (quetiapine XR 150 mg/d); neither was considered treatment related by the investigator.
The overall incidence of AEs in the 6-wk randomized phase was 66.9%, 82.4%, and 87.2% in the placebo, quetiapine XR 150 mg/d, and 300 mg/d groups, respectively; the majority of these were mild to moderate in intensity. The most commonly reported AEs (>5%) are given in Table 4.
Patients with multiple events falling under the same preferred term are counted only once in that term. AEs reported as more than one AE could be experienced by one patient.
Somnolence and sedation
Overall, the incidence of AEs related to somnolence (MedDRA-preferred terms somnolence, sedation, lethargy, and sluggishness) was 8.8%, 46.6%, and 50.3% for placebo, quetiapine XR 150 mg/d and 300 mg/d, respectively. AEs related to somnolence were generally mild to moderate in intensity and most did not lead to withdrawal from the study. The median time to onset of AEs associated with somnolence was 8 d in the placebo group compared to 2 d in both quetiapine XR groups. The mean duration of AEs associated with somnolence was similar in the placebo, quetiapine XR 150 mg/d, and 300 mg/d groups at 24, 27, and 23 d, respectively.
The incidence of AEs potentially related to EPS (MedDRA-preferred terms akathisia, restlessness, tremor, extrapyramidal disorder, psychomotor hyperactivity, hypertonia, drooling, and cogwheel rigidity) was 3.4% in the placebo and quetiapine XR 150 mg/d groups, and 8.1% in the quetiapine XR 300 mg/d group. The two most commonly reported AEs related to EPS were akathisia (0.7%, 1.4%, and 2.7%) and restlessness (1.4%, 0.7%, and 2.7%) in the placebo, quetiapine XR 150 mg/d, and 300 mg/d groups, respectively. At study end, the majority of patients in all treatment groups experienced ‘no change’/‘improvement’ in their SAS total (89.5–95.5%) and BARS global (95.6–99.3%) scores.
The incidence of AEs potentially related to sexual dysfunction (MedDRA-preferred terms libido decreased, libido increased, and sexual dysfunction) was: 0.7%, placebo; 0.7%, quetiapine XR 150 mg/d; and 2.0%, quetiapine XR 300 mg/d. All of these AEs were of mild to moderate intensity and did not lead to discontinuation.
Vital signs and laboratory parameters
At the end of the study, there were no notable differences in the mean changes from baseline to the end of treatment in vital signs (including orthostatic changes), ECG, or haematology data between either dose of quetiapine XR and placebo. Mean (s.d.) changes in orthostatic, supine, and standing pulse, respectively, at the end of treatment were −0.1 (7.9), 1.8 (9.3), and 1.7 (10.7) b.p.m. with quetiapine XR 150 mg/d; 0.3 (8.3), 3.5 (11.7), and 3.8 (12.7) b.p.m. with quetiapine XR 300 mg/d; and 0.5 (7.4), −0.1 (10.8), and 0.4 (11.7) b.p.m. with placebo; changes appeared to be treatment related. A potentially clinically important shift (⩾60 ms) in QTc interval occurred in one patient in the quetiapine XR 150 mg/d treatment group.
The mean changes from baseline to end of treatment for glucose and lipid laboratory parameters and body weight, and clinically relevant shifts in these parameters, are given in Table 5.
AD, Antidepressant; HDL, high-density lipoprotein; LDL, low-density lipoprotein.
Fasting status was determined based upon a documented report from the patient that last meal was ⩾8 h before blood sample taken for randomization and post-randomization laboratory measurements. However, not all samples could be confirmed as fasted despite there being an 8-h interval since the last meal, as patients could have had caloric intake.
The n values are for the complete safety analysis set; numbers of patients varied for each parameter.
Patients with fasting status assumed.
One patient in the quetiapine XR 150 mg/d group (0.7%) had an AE potentially related to suicidality. This AE of mild suicidal ideation occurred on day 4, lasted for 5 d, and was not reported as a serious AE. The patient had a history of two suicide attempts prior to enrolment in the study and subsequently discontinued from the study due to an AE of hypersomnia.
The percentage of patients with a MADRS item 10 (suicidal thoughts) score ⩾4 at any time during the study was 0.7% (placebo), 1.4% (quetiapine XR 150 mg/d), and 0.7% (quetiapine XR 300 mg/d). From the Columbia-like suicidality analysis, 3.4% and 1.3% of patients in the quetiapine XR 150 mg/d and 300 mg/d groups, respectively, experienced events classified as ‘other’. In the placebo group, ‘suicidal ideation’ (0.7%), ‘self-injurious behaviour, intent unknown’ (0.7%), ‘not enough information, non-death’ (0.7%), and ‘other’ (1.4%) events were reported following Columbia-like suicidality analysis.
2-wk drug-discontinuation phase
Of the patients who completed the 6-wk randomized treatment phase, 79.2%, 80.7%, and 64.8% of patients in the placebo, quetiapine XR 150 mg/d, and 300 mg/d groups, respectively, completed the 2-wk drug-discontinuation phase.
During the 2-wk drug-discontinuation/follow-up phase, the most common AEs (>5%) were insomnia and headache in the quetiapine XR 150 mg/d group, and nausea and insomnia in the quetiapine XR 300 mg/d group. No AEs were reported by >5% of patients in the antidepressant alone group.
The most common reasons for discontinuation during the drug-discontinuation phase, apart from ‘other’, were: ‘patient did not complete day 14 TDSS assessment’, ‘patient not willing to continue’, and ‘patient lost to follow-up’.
Abrupt discontinuation of study drug at the end of the randomized phase resulted in higher TDSS total scores for patients who had received quetiapine XR than for those who had received placebo; most relative elevations occurred in the first week following discontinuation. TDSS total mean scores (minimum and maximum values) over the discontinuation period were: 1.9–3.1 (placebo), 2.4–4.6 (quetiapine XR 150 mg/d), and 2.2–4.5 (quetiapine XR 300 mg/d) where higher scores indicate a greater number of reported TDSS symptoms. The most pronounced effects were seen for symptoms of insomnia, sweating, chills, headache, and nausea.
This large-scale, randomized, placebo-controlled study evaluated adjunctive quetiapine XR 150 mg/d and 300 mg/d in patients with MDD who had an inadequate response to ongoing antidepressant treatment. Results from a variety of assessment measures, including MADRS, HAMD, HAMA, CGI-S, and CGI-I rating scales, show that quetiapine XR 300 mg/d improved symptoms of depression and anxiety in patients with MDD. For quetiapine XR 150 mg/d, changes in the majority of these measures did not reach statistical significance vs. placebo.
Reduction of depressive symptoms with quetiapine XR was observed at week 1, with significant improvements in MADRS total scores. Early symptom relief may contribute to improved treatment adherence as patients are more likely to continue with a particular treatment they feel is effective (Barbee et al.2004). Moreover, inner tension, sleep disturbances, and pessimistic thoughts are common features of clinical depression, and as such, the efficacy of quetiapine XR 300 mg/d across these domains, plus the significant reduction in MADRS item 10 (suicidal thoughts) scores, suggest that adjunct quetiapine XR 300 mg/d may prove beneficial to patients who have not achieved an adequate response with their ongoing antidepressant.
It should be noted that early symptom improvement was maintained to week 6 for the 300 mg/d quetiapine XR dose and not for the 150 mg/d dose. These results are in contrast to an almost identical previous study of adjunctive quetiapine XR in patients with MDD and an inadequate response to their ongoing antidepressant, in which both 150 and 300 mg/d quetiapine XR significantly reduced MADRS total scores at weeks 1, 2, and 4, and at week 6 (primary endpoint) (Bauer et al.2009). Although both studies enrolled patients with an inadequate response to ongoing antidepressant treatment, differences between patient populations in terms of chronicity of depression may account for the contrast in study outcomes for the quetiapine XR 150 mg/d dose. In the present study, a higher percentage of patients had recurrent MDD (90.4–94.4%) compared to those in the Bauer et al. study (80.6–82.0%). A history of recurrent episodes has been associated with failure to respond to two adequate trials of different antidepressants (Souery et al.2007). Overall, in patients with MDD and an inadequate response to ongoing antidepressant, quetiapine XR dosed at 300 mg/d may be more effective than at 150 mg/d.
The a priori remission criterion in this study was MADRS total score ⩽8; however, remission using criteria of MADRS total scores ⩽10 or ⩽12 was also assessed. Regardless of the criteria used, a greater proportion of patients experienced remission at week 6 with quetiapine XR 300 mg/d compared to placebo; however, the difference with quetiapine XR 150 mg/d was not statistically significant. In addition, in this study, patients treated with quetiapine XR 300 mg/d achieved significant improvements from randomization in HAMA total scores at week 6; it may be suggested that relieving the residual anxiety symptoms and suicidal thoughts contributed to achieving higher remission rates with quetiapine XR 300 mg/d compared to 150 mg/d.
The efficacy of aripiprazole and olanzapine as augmentation agents for treatment-resistant depression has also been investigated (Berman et al.2007, 2009; Marcus et al.2008; Thase et al.2007). Although varying criteria were used, remission rates of 25.4–36.8% were reported for these agents. As the present study population had a history of inadequate response to ongoing antidepressant therapy, our patients may have been less treatment resistant than those enrolled in the aripiprazole or OFC studies; remission rates with quetiapine XR 300 mg/d were higher than those observed with aripiprazole or OFC regardless of the criteria used to define remission.
The patient population in the present study had an inadequate response to ongoing antidepressant treatment; following 6 wk augmentation with quetiapine XR 150 and 300 mg/d, and placebo, response rates were similar to those in the almost identical Bauer et al. (2009) study (55.4%, 57.8%, and 46.3% for quetiapine XR 150 and 300 mg/d and placebo, respectively). Response rates of 46.6% and 26.6% were reported for aripiprazole and placebo adjunct to ongoing antidepressants, respectively, in patients with treatment-resistant MDD (Berman et al.2009). Although these inter-study differences in response rates may be due to more or less treatment-refractory patient populations, quetiapine XR 300 mg/d demonstrated statistical significance vs. placebo in both the present study and the Bauer et al. (2009) study, despite the relatively high placebo response.
The mode of action of quetiapine has not been fully understood; however, the recent characterization of the major active human metabolite following treatment with quetiapine, N-desalkylquetiapine (norquetiapine), has provided a potential explanation for the antidepressant effects seen in clinical trials (Jensen et al.2008). Both quetiapine and norquetiapine have moderate-to-high affinity for serotonin 5-HT2A and dopamine D2 receptors; norquetiapine is also a potent inhibitor of the norepinephrine transporter (NET) (Goldstein et al.2008; Jensen et al.2008). Evidence for the clinical relevance of these findings has been supported by positron emission tomography (PET) imaging of NET occupancy in quetiapine-treated subjects (Nyberg et al.2008). NET inhibition has not been shown by other atypical antipsychotics at clinically relevant doses; however, it is a property shared by a number of traditional antidepressant therapies, such as SNRIs, and may contribute to the antidepressant effect. In addition, clinical experience suggests that simultaneous targeting of both the noradrenergic and serotonergic systems is one of the most effective augmentation strategies in patients with partial response or non-response to antidepressant therapy (Hirschfeld et al.2002).
In this study, safety and tolerability results with quetiapine XR adjunct to antidepressant treatment were consistent with the known profile of quetiapine in other indications, with the most common AEs being dry mouth, somnolence, sedation, dizziness, constipation, and fatigue (AstraZeneca, 2008). The most common reason for discontinuing from the study was due to AEs, the incidence of which was highest in the quetiapine XR 300 mg/d group, followed by the quetiapine XR 150 mg/d group, and the placebo group. These data are in keeping with those reported by Bauer et al. (2009) . Overall, discontinuations due to AEs were generally reported as related to somnolence or sedation; however, most AEs of somnolence did not lead to withdrawal from the study. In the present adjunct study, the overall incidence of AEs potentially related to EPS was 3.4%, 8.1%, and 3.4% in the quetiapine XR 150 mg/d and 300 mg/d, and placebo groups, respectively; the incidence of akathisia was 1.4%, 2.7%, and 0.7%, respectively. Recent studies of adjunctive aripiprazole vs. placebo in patients with MDD reported EPS rates of 27.5% vs. 9.7%, respectively (Berman et al.2007), and akathisia rates of 18.2% vs. 3.5%, respectively (Berman et al.2009). The overall EPS rates in several short-term studies of OFC in patients with treatment-resistant depression have not been reported; however, tremor occurred with an incidence of 10.5–11.6%, 2.1–8.7%, and 4.9–8.0% for OFC, fluoxetine, and olanzapine, respectively (Shelton et al.2005; Thase et al.2007). For quetiapine XR adjunct to ongoing antidepressant therapy, tremor was not reported by >5% of patients in this study. While atypical antipsychotics are associated with a lower risk for EPS and tardive dyskinesia than conventional antipsychotics, it is important that patients receiving atypical antipsychotics are monitored for the emergence of events potentially related to EPS (Casey, 2006). As this was a short-term study, occurrences of tardive dyskinesia would not be expected and there were no reports of this AE during the study.
Consistent with findings from the previous adjunct study (Bauer et al.2009), adjunctive quetiapine XR was associated with greater weight gain (weight changes in the present study were 0.8, 1.6, and 0.3 kg with quetiapine XR 150 and 300 mg/d, and placebo, respectively) and a higher proportion of patients with clinically relevant shifts in weight (1.4%, 7.6%, and 2.1%, respectively) than placebo. Other studies with adjunctive atypical antipsychotics have also reported weight gain. In patients with treatment-resistant MDD, weight changes were 1.2 and 0.8 kg with adjunct aripiprazole and placebo (Berman et al.2009), and 4.9, 0.4, and 5.5 kg with OFC, fluoxetine, and olanzapine (Thase et al.2007), respectively. Clinically relevant shifts in body weight occurred in 4.5% and 1.2% of patients treated with adjunct aripiprazole and placebo. The proportion of patients experiencing clinically relevant shifts in body weight was not reported for OFC; however, 35.0%, 6.8%, and 39.7% of patients in the OFC, fluoxetine, and olanzapine groups, respectively, reported weight gain as an AE.
In both quetiapine XR treatment groups, changes from randomization were also observed in fasting glucose and fasting total, high-density lipoprotein and low-density lipoprotein cholesterol values. These changes in laboratory parameters and weight are consistent with observations in short-term monotherapy studies from the quetiapine XR clinical programme in patients with MDD (Cutler et al.2009; Weisler et al.2009). Publication of the safety and tolerability results from a longer-term monotherapy study in patients with MDD is awaited. The potential for AEs and differences in tolerability profiles between augmentation agents should be considered before initiating any treatment regimen in patients with MDD.
A strength of this study is the large patient population. The relative distribution of ongoing antidepressants used in our study population is similar to the overall pattern of antidepressants prescribed in clinical practice and mimics a ‘real-life’ situation where patients may be receiving any available antidepressant drug. All three treatment groups had similar proportions of patients receiving each antidepressant [agent and class of agent (SSRI, SNRI, other)]. Antidepressant doses and prior treatment durations were also similar across randomized groups, therefore, neither ongoing nor prior antidepressant treatment are considered to have influenced the results of this study.
A limitation of this study was that dosing was fixed, which is not reflective of clinical practice where clinicians can alter dose to optimize efficacy or reduce treatment-related AEs. It is possible that some patients with MDD in this study may have received a dose of quetiapine XR that was either too high or too low; therefore, flexible-dosing studies would be of value in order to fully characterize the optimal dose range. The short-term nature of this study means that these results cannot be extrapolated to the longer term.
In summary, this double-blind study conducted in a large population of patients with MDD who had an inadequate response to ongoing antidepressant treatment provides further evidence for the effectiveness of quetiapine XR, with efficacy confirmed using a variety of clinical rating scales. Here we report that quetiapine XR 150 mg/d as adjunct therapy was not statistically different from placebo at week 6, despite demonstrating efficacy at week 1. Quetiapine XR 300 mg/d was effective against a range of depressive symptoms at week 6 in this patient population, with onset of efficacy seen at week 1. Both doses of quetiapine XR studied were generally well tolerated.
This study was funded by AstraZeneca, manufacturer of quetiapine XR. We thank Dr Kim Croskery from Complete Medical Communications, who provided medical writing support funded by AstraZeneca. The following investigators were involved in this study (Study 6, Pearl): Amit Anand (Indianapolis, Indiana), Sarah Atkinson (Rochester, New York), Michael Banov (Roswell, Georgia), Benny Barnhart (Wichita Falls, Texas), Brian Bortnick (Atlanta, Georgia), Ronald Brenner (Cedarhurst, New York), Edward Cherlin (El Centro, California), Adnan Dahdul (Springfield, Massachusetts), Nizar El-Khalili (Lafayette, Indiana), Prakash Ettigi (Richmond, Virginia), Miguel Flores (Hialeah, Florida), Richard Jackson (Royal Oaks, Michigan), Elias Sarkis (Gainsville, Florida), Anita Kablinger (Shreveport, Louisiana), James Kocsis (New York, New York), Jelena Kunovac (Oceanside, California), Charles Morin (Braintree, Massachusetts), Amy Mulroy (San Antonio, Texas), Veronique Sebastian (Oklahoma City, Oklahoma), Ismail Sendi (Clinton, Michigan), Phebe Tucker (Oklahoma City, Oklahoma), Robert Buynak (Valparaiso, Indiana), Carlos Danger (Miami, Florida), Kettlie Daniels (Toledo, Ohio), Robert Earle (Friendswood, Texas), Sanjay Gupta (Olean, New York), Gregory Mattingly (St Charles, Missouri), Haydn Thomas (Prairie Village, Kansas), Ethan Kass (Coral Springs, Florida), James Whalen (Lincoln, Rhode Island), Jeffrey Ross (Hoffman Estates, Illinois), Jerold Kreisman (St Louis, Missouri), Barbara Harris (Phoenix, Arizona), Joseph Ripperger (Norman, Oklahoma), Michael Levy (Staten Island, New York), Lora McGill (Memphis, Tennessee), Mark Joyce (Jacksonville, Florida), Charles Bailey (Orlando, Florida), Charles Meredith (San Diego, California), Ramanath Gopalan (Arlington, Virginia), Riaz Baber (Naperville, Illinois), Bijan Bastani (Beechwood, Ohio), Elly Lee (Irvine, California), Robert Hudrick (Cherry Hill, New Jersey), Deborah Bergen (Wichita, Kansas), Susanna Goldstein (New York, New York), Robert Riesenberg (Atlanta, Georgia), Fares Arguello (Salt Lake City, Utah), David Krefetz (Clementon, New Jersey), Steven Eisen (Philadelphia, Pennsylvania), Mohammed Alam (Oak Brook, Illinois), Arifulla Khan (Bellevue, Washington). [Clinical trials registry number: NCT00326105.]
Statement of Interest
Nizar El-Khalili has received research grants from AstraZeneca, Eli Lilly, GlaxoSmithKline, Pfizer, and Sanofi-aventis, and has participated in advisory boards for Eli Lilly and speakers' bureaux for AstraZeneca, Eli Lilly, and Sanofi-aventis. Mark Joyce has worked on clinical trials for the following companies: Abbott, Addrenex Pharma, Allergan, Allon Therapeutics, AstraZeneca, Avera, Bristol–Myers Squibb, Cephalon, DOV, Eli Lilly, Epix Pharmaceuticals, Forest, GlaxoSmithKline, Labopharm, Merck, McNeil, Myriad, Neurocrine, New River Pharmaceuticals, Novartis, Ono Pharma USA, Organon, Pfizer, Sanofi, Sepracor, Shire, Solvay, Somaxon, TransTech, and Wyeth. Sarah Atkinson has received research support from: Eli Lilly, AstraZeneca, Bristol–Myers Squibb, Insys, Lundbeck, Pfizer, Shire Pharmaceuticals, Sanofi-aventis, and Takeda Pharmaceuticals. Robert Buynak has received research support from AstraZeneca. Catherine Datto, Petter Lindgren, and Hans Eriksson are employees of AstraZeneca.