George I. Papakostas, MD

Despite the progressive increase in the number of available antidepressants, many patients with depression continue to be symptomatic. For example, it was reported recently that as many as half of the patients enrolled in 2 academic-based depression specialty clinics did not achieve remission despite receiving numerous antidepressant treatment trials.

 To complicate matters further, residual symptoms among remitters are common and are associated with poorer psychosocial functioning as well as increased relapse rates. In light of the challenge that treatment-resistant major depressive disorder (TRD) poses for clinicians and patients alike, there is an urgent need to develop novel treatment strategies for resistant depression that are safer and more effective than those currently in use.

In 2006, there have been two areas of important development with regard to TRD: the Sequential Treatment Alternatives to Relieve Depression (STAR*D) trial and the use of the atypical antipsychotic agents as an adjunct to standard antidepressant treatment. In this article, I will provide an overview of the definition, prevalence, and staging of TRD and then discuss published findings from the STAR*D study as well as recent developments regarding the use of atypical antipsychotics in TRD.

Definition, prevalence, and staging of TRD

TRD typically refers to an inadequate response to at least one antidepressant trial of adequate dosage and duration in patients with unipolar depressive disorders. Adequate duration is often defined as a minimum of 6 weeks, stemming from the observation that fewer than 7% of patients who show little improvement following 6 weeks of treatment with fluoxetine eventually respond (ie, show a 50% decrease in symptom severity) following an additional 2 weeks of treatment.

 Although similar analyses of clinical trials of longer duration (12 weeks) have provided evidence that 6 weeks may be too short a duration to declare an antidepressant trial ineffective,

 it is important to remember that spontaneous remission can occur over time. The definition of adequate dosing varies widely from agent to agent, with values deriving from double-blind placebo-controlled trials or dose-comparator studies (ie, the lowest dosage at which efficacy was demonstrated in double-blind placebo-controlled trials).

Definitions of "adequate response" have varied throughout the past few decades, ranging from the more traditional view in which treatment resistance is defined as strict nonresponse (25% or less improvement in symptoms following treatment) to the broadest definition (failure to achieve full remission of the depressive episode). Nowadays, most experts agree for several reasons, that inadequate response is the failure to achieve full symptom remission.

 First of all, patients presenting with moderate to severe depression may still be quite symptomatic despite a 25% to 50% improvement in depressive symptoms. In addition, residual symptoms have been associated with poorer psychosocial functioning,

 Finally, incomplete response (defined as a 25% or greater improvement in depressive symptoms while failing to achieve remission) appears to be more than twice as common as strict nonresponse in naturalistic treatment settings.

In recent years, increasing attention has also been paid to developing methods to stage the degree of resistance in patients with unipolar depressive disorders. Thase and Rush

 first proposed a 5-stage model that provides a categoric assignment of degree of resistance. According to this model, failure of a trial of one major class of antidepressants constitutes stage 1 antidepressant resistance; failure of an additional antidepressant trial from a different class constitutes stage 2 resistance; failure of an additional tricyclic antidepressant (TCA) trial, stage 3 resistance; and failure of an additional monoamine oxidase inhibitor (MAOI) trial, stage 4 resistance. Failing a trial of electroconvulsive therapy (ECT) in addition to those treatments cited in stages 1 through 4 has been defined as stage 5 treatment resistance, according to this model.

Several methodologic issues have been raised concerning these stages: (1) that the degree of intensity of each trial in terms of dosing and duration is not accounted for, (2) that the model assumes that depression that does not respond to 2 agents of different classes is more difficult to treat than depression that fails to respond to 2 agents of the same class, and (3) that the role of augmentation or combination strategies is not considered. An additional limiting factor is the implicit hierarchy of treatments, with MAOIs considered superior to TCAs.

More recently, the Massachusetts General Hospital staging method was proposed.

 According to this model, nonresponse to each adequate trial increases the overall degree of resistance by 1 point, while optimization of dosage and duration, augmentation, or combination increases the degree of resistance by 0.5 points. ECT was proposed to increase the degree of resistance by 3 points according to this model. Recently, empiric testing of these 2 models with the use of outcome data from outpatients receiving treatment for depression in 1 of 2 hospital-based, academically affiliated depression specialty clinics revealed that, although highly correlated, the latter model demonstrated significantly greater ability to predict remission status than the Thase-Rush model did.

Estimates regarding the prevalence of treatment-resistant depression are available from several sources. Fava and Davidson

 reviewed 36 clinical trials (both open-label and placebo-controlled) of antidepressants for unipolar depression involving more than 3500 patients and reported that between 29% and 46% of patients failed to respond fully following a single antidepressant treatment. Corey-Lisle and colleagues

 reported that symptoms in approximately 22% of patients who received treatment for depression from their primary care physicians remitted following 6 months of treatment; 32% of the patients were partial responders, while 45% were nonresponders. Similarly, Rush and colleagues

 reported an 11% remission rate and a 26.3% response rate in outpatients with depression following 12 months of treatment in 1 of several public-sector community clinics. Finally, Petersen and colleagues

 reported a 50.4% remission rate in outpatients with major depressive disorder (MDD) enrolled in 1 of 2 depression specialty clinics.

Management of treatment-resistant depression

For patients who continue to experience symptoms of depression despite an optimal antidepressant trial, 2 general treatment strategies exist. The first strategy involves switching from one antidepressant to another. The second strategy, termed augmentation, involves adding an antidepressant or nonantidepressant agent to the treatment regimen. Despite the prevalence and consequences of TRD, and despite decades of clinical trials focusing on exploring effective and safe treatments for resistant depression, no gold-standard treatment or treatment consensus among experts has emerged, a fact largely due to the serious lack of positive results in double-blind placebo-controlled studies. In fact, the majority of double-blind augmentation studies conducted that focused on a handful of compounds have yielded discouraging results. Such adjunctive compounds have included (but were not limited to) lithium,

 In addition, small placebo-controlled studies that identify promising compounds for use as adjunctive treatment in TRD have often gone unreplicated.

Even less is known regarding the relative efficacy of the various possible switching strategies (ie, from an SSRI to either an SSRI, a serotonin-norepinephrine reuptake inhibitor, a norepinephrine-dopamine reuptake inhibitor, or vice versa). As a result, more often than not, clinicians' decisions regarding the treatment of resistant depression are guided by anecdotal reports, case series, uncontrolled trials or, at best, small and unreplicated double-blind studies. However, in light of the challenge TRD poses for clinicians and patients alike, identification of novel treatments for resistant depression is urgently needed to help further refine the standard of care for depression. In 2006, there were 2 major areas of interest regarding TRD: the STAR*D trial and the use of atypical antipsychotic agents as an adjunct to standard antidepressant treatment.

STAR*D is a multisite, prospective, randomized, multistep clinical trial of outpatients with nonpsychotic MDD. The study compared various treatment options for those who did not attain a satisfactory response with citalopram, an SSRI. The study enrolled 2876 adults with MDD (aged 18 to 75 years) from 18 primary care and 23 psychiatric practices. After receiving open-label citalopram (20 to 60 mg/d) for up to 14 weeks, participants without sufficient symptomatic benefit were eligible for randomization to level 2 treatments. Level 2 involved the use of 3 switch (sertraline, bupropion, venlafaxine) and 2 augmentation (bupropion, buspirone) pharmacotherapies. Patients who did not experience sufficient symptomatic benefit were then eligible for random assignment to 2 switch (mirtazapine or nortriptyline) or 2 augmentation (lithium or thyroid hormone) treatments in level 3. Finally, those without sufficient improvement at level 3 were eligible for random assignment to 1 of 2 switch options--tranylcypromine or the combination of mirtazapine and venlafaxine. 

The primary outcome for the entire STAR*D trial was remission, defined as a 17-item Hamilton Depression Rating Scale (HDRS-17) score of less than 8 at study end point. To date, results have been published for level 1,

). Overall HDRS-17 remission rates for level 1 were 28%. This result is highly informative, since it demonstrates that remission rates were similar between the STAR*D study sample, a highly generalizable sample with substantial Axis I and Axis III comorbidity, and patients typically enrolled in randomized clinical trials. The difference in remission rates between the level 2 augmentation treatments (bupropion and buspirone) was not statistically significant (29.7% vs 30.1%). However, fewer patients treated with adjunctive bupropion discontinued treatment because of intolerance than patients treated with adjunctive buspirone (12.5% vs 20.6%; 

 < .009), suggesting that while both treatment approaches appeared to be useful, there was an advantage for bupropion augmentation in terms of tolerability.

There were also no differences in remission rates between the level 2 switch treatments (24.8%, 21.3%, and 17.6% for venlafaxine, bupropion, and sertraline, respectively), suggesting that all 3 medications were a reasonable choice for patients with SSRI-resistant depression. Finally, there was no statistically significant difference in remission rates between level 3-treatment patients randomized to switch to either mirtazapine or the TCA nortriptyline (12.3% vs 19.8%, respectively). Notable in the level 3 study results were the very low remission rates obtained when patients who had failed to respond to 2 adequate treatments were switched to a third (fewer than 20% in both cases).

Although, to date, the STAR*D trial has failed to identify a treatment strategy that is clearly superior in terms of its overall efficacy and tolerability, the STAR*D database will allow clinicians to examine whether any of these tested treatments are particularly effective among specific subpopulations of patients with MDD (termed "MDD subtypes"), defined either by the presence (or absence) of clinical symptoms, syndromes, comorbid disorders, or genotype. Hypotheses generated from this invaluable database could then be tested prospectively in future randomized, double-blind, placebo-controlled trials.

Augmentation with atypical antipsychotic agents

Atypical antipsychotics are a heterogeneous group, each with a distinct and complex set of receptor affinities involving dopaminergic and serotonergic receptors as well as various effects on noradrenergic, histaminergic, and cholinergic systems. As a result, the receptor-binding profile of the atypical neuroleptics differs substantially from that of the typical antipsychotics. This difference may provide clinical advantages to the atypical antipsychotic agents in therapeutic areas other than schizophrenia.

Specifically, similar to the typical antipsychotic agents, the atypical antipsychotics act as dopamine-2 (D

 However, unlike typical antipsychotics, the atypical antipsychotics act, with varying degrees of affinity, as serotonin-2 (5-HT

) receptor antagonists, with ziprasidone, aripiprazole, and risperidone also possessing affinity for the serotonin-1A (5-HT

) receptors. This is of particular interest because agents that act as agonists or antagonists of the 5-HT

 receptor (including pindolol, buspirone, gepirone, ipsapirone, zalospirone), as well as agents that are selective 5-HT

receptor antagonists (including trazodone and nefazodone) have shown antidepressant effects in clinical trials.

 Aripiprazole and ziprasidone also have additional unique neurochemical properties that distinguish them from the other atypicals. Aripiprazole, for instance, is a potent D

receptor partial agonist, while ziprasidone has been shown to inhibit the neuronal uptake of serotonin and norepinephrine.

Although the results of several open-label trials suggested a potential role for the atypical antipsychotics in TRD,

 there has been a paucity of double-blind placebo-controlled studies confirming that this treatment strategy is truly effective. Before 2006, for instance, only 3 double-blind placebo-controlled trials focusing on the use of adjunctive atypical antipsychotics for TRD had been published or presented at a major psychiatric meeting.

In the first such study, Shelton and colleagues

 randomly selected 28 outpatients who had fluoxetine-resistant MDD to continue treatment with fluoxetine monotherapy, undergo a switch to olanzapine monotherapy, or receive olanzapine augmentation for 8 weeks. Greater symptom resolution was reported for the combination group than for either monotherapy group.

 also focusing on the use of adjunctive olanzapine in TRD, failed to confirm these findings.

 In 2006 alone, however, 7 new double-blind placebo-controlled trials focusing on the use of adjunctive olanzapine,

 were presented at 2 major scientific meetings. All 7 studies reveal greater efficacy for adjunctive treatment with an atypical antipsychotic agent than for placebo in TRD.

) shows a brief overview of studies using atypical antipsychotics for adjunctive therapy in TRD. The results from these studies appear to be very promising for the use of this treatment strategy. However, although it appears that the sum of the evidence supports a potential role for the atypical antipsychotic agents as adjuncts in TRD, the evidence is far from definitive. Specifically, 2 important limitations remain. First, the extent to which the pooled findings apply to each of the 5 atypical antipsychotic agents (risperidone, olanzapine, quetiapine, ziprasidone, and aripiprazole) remains undetermined, and there appears to be significant variability in the evidence stemming from each of these 5 agents. For example, none of the studies involved the use of either ziprasidone or aripiprazole. A second limitation is that all 10 clinical trials were funded by the makers of the atypical antipsychotic agents, while none of the studies were funded by independent sources (ie, government sources or private not-for-profit organizations). This may have introduced bias in the meta-analysis.

Therefore, in light of the challenge that TRD poses to clinicians and patients alike, independently funded studies establishing the efficacy, safety, and tolerability of each individual antipsychotic for treatment of TRD are warranted. If they are safe and effective as antidepressant adjuncts, the atypical antipsychotic agents would represent an attractive option for many patients who have had unsatisfactory initial response to standard treatment. If not found to be either safe or effective, this would also be highly informative given the significant proportion of TRD patients for whom, despite the relative paucity of data from studies of rigorous design, atypical antipsychotic agents are prescribed off-label.

There is an urgent need to develop novel treatment strategies for resistant depression that are both safer and more effective than those currently employed. Fortunately, 2006 has seen 2 major developments for TRD. The results of STAR*D published thus far suggest that popular treatment choices for TRD are, more or less, equally effective. Sub- sequent exploratory analyses of this database may provide invaluable hints about whether it is possible to begin to "match" any of these tested treatments with specific subpopulations of patients with MDD (MDD subtypes). Hypotheses generated from this invaluable database could then be tested prospectively in future randomized, double- blind, placebo-controlled trials. Finally, rapidly accumulating data from double-blind placebo-controlled trials suggest that augmentation of standard antidepressants with atypical antipsychotic agents, a relatively new treatment strategy, may also be effective for patients with MDD who do not experience sufficient symptom improvement despite adequate treatment with antidepressants. Despite these 2 developments, however, further controlled clinical trials are essential to identify the most effective treatment strategies for TRD.

Dr Papakostas is assistant professor of psychiatry at Harvard Medical School and a staff psychiatrist at Massachusetts General Hospital in Boston. He reports that he has served as a consultant for the Aphios Corporation, GlaxoSmithKline, Evotec, Inflabloc Pharmaceuticals Inc, Jazz Pharmaceuticals, Pamlab LLC, and Pfizer Inc. He has received honoraria from Evotec, GlaxoSmithKline, Inflabloc Pharmaceuticals Inc, Jazz Pharmaceuticals Inc, Pamlab LLC, Pfizer Inc, and Titan Pharmaceuticals, and he has received research support from Bristol-Myers Squibb, Pamlab LLC, and Pfizer Inc.

 Petersen T, Papakostas GI, Posternak MA. Empirical testing of two models for staging antidepressant treatment resistance. 

 Nierenberg AA, McLean NE, Alpert JE, et al. Early nonresponse to fluoxetine as a predictor of poor 8-week outcome.

 Quitkin FM, Petkova E, McGrath PJ, et al. When should a trial of fluoxetine for major depression be declared failed?

 Fava M. Diagnosis and definition of treatment-resistant depression. 

 Papakostas GI, Petersen T, Denninger JW, et al. Psychosocial functioning during the treatment of major depressive disorder with fluoxetine.

 Paykel ES, Ramana R, Cooper Z, et al. Residual symptoms after partial remission: an important outcome in depression. 

 Petersen T, Papakostas GI, Posternak MA, et al. Empirical testing of two models for staging antidepressant treatment resistance. 

 Thase ME, Rush AJ. When at first you don't succeed: sequential strategies for antidepressant nonresponders. 

 Fava M, Davidson KG. Definition and epidemiology of treatment-resistant depression. 

 Corey-Lisle PK, Nash R, Stang P, Swindle R. Response, partial response, and nonresponse in primary care treatment of depression. 

 Rush AJ, Trivedi M, Carmody TJ, et al. One-year clinical outcomes of depressed public sector outpatients: a benchmark for subsequent studies. 

 Fava M, Alpert J, Nierenberg A, et al. Double-blind study of high-dose fluoxetine versus lithium or desipramine augmentation of fluoxetine in partial responders and nonresponders to fluoxetine. 

 Aronson R, Offman HJ, Joffe RT, Naylor CD. Triiodothyronine augmentation in the treatment of refractory depression: a meta-analysis. 

Ballesteros J, Callado LF. Effectiveness of pindolol plus serotonin uptake inhibitors in depression: a meta-analysis of early and late outcomes from randomised controlled trials. 

 Carpenter LL, Yasmin S, Price LH. A double-blind, placebo-controlled study of antidepressant augmen- tation with mirtazapine. 

Ferreri M, Lavergne F, Berlin I, et al. Benefits from mianserin augmentation of fluoxetine in patients with major depression non-responders to fluoxetine alone. 

 Pope HG Jr, Cohane GH, Kanayama G, et al. Testosterone gel supplementation for men with refractory depression: a randomized, placebo-controlled trial. 

 Trivedi MH, Rush AJ, Wisniewski SR, et al. Evaluation of outcomes with citalopram for depression using measurement-based care in STAR*D: implications for clinical practice. 

Rush AJ, Trivedi MH, Wisniewski SR, et al. Bupropion-SR, sertraline, or venlafaxine-XR after failure of SSRIs for depression. 

 Trivedi MH, Fava M, Wisniewski SR, et al. Medication augmentation after the failure of SSRIs for depression. 

 Fava M, Rush AJ, Wisniewski SR, et al. A comparison of mirtazapine and nortriptyline following two consecutive failed medication treatments for depressed outpatients: a STAR*D report. 

 Papakostas GI, Petersen TJ, Nierenberg AA, et al. Ziprasidone augmentation of selective serotonin reuptake inhibitors (SSRIs) for SSRI-resistant major depressive disorder. 

 Papakostas GI. Augmentation of standard antidepressants with atypical antipsychotic agents for treatment-resistant major depressive disorder. 

 Papakostas GI, Petersen TJ, Kinrys G, et al. Aripiprazole augmentation of selective serotonin reuptake inhibitors for treatment-resistant major depressive disorder.

 Shelton RC, Tollefson GD, Tohen M, et al. A novel augmentation strategy for treating resistant major depression. 

Articles

Despite the progressive increase in the number of available antidepressants, many patients with depression continue to be symptomatic.

George I. Papakostas, MD

Articles

Treatment-Resistant Major Depressive Disorder: New Developments in 2006