Recent developments pertaining to treatment‐resistant depression: a 40‐year perspective

With the increasing recognition that major depressive disorder (MDD) is one of the world greatest public health problems1, 2, there have recently been concerted efforts to ensure that people suffering from this condition are promptly recognized, accurately diagnosed, and vigorously treated. Indeed, a relatively wide range of proven treatments are now available to help depressed people, and health care systems and agencies throughout the world have prioritized implementation strategies to efficiently deliver cost-effective interventions1. Without established primary prevention strategies to reduce the incidence of MDD, maximizing access to treatment and ensuring optimal delivery of care represents the best way to reduce the morbidity, mortality, and personal and societal costs of this common condition1. That said, no more than one half of people who receive an adequate course of a first-line antidepressant medication will obtain an acceptable response (i.e., at least a 50% reduction in depressive symptom severity, coupled with a tolerable level of side effects) and, as illustrated by the results of the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study3, the likelihood of benefit diminishes substantially after the second sequential treatment trial. Episodes that follow this course are commonly called treatment-resistant depression (TRD), and account for a disproportionately large proportion of the illness burden associated with MDD1. This is the subject of the excellent paper by McIntyre et al4, which provides a concise, yet comprehensive review of the topic, including up-to-date summaries of the best studied and most promising treatment strategies. The concept of TRD is nearly as old as the first generation of effective treatments for depression, namely electroconvulsive therapy (ECT), tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs). The first papers using that concept were published in the 1970s5. In that era, an almost intuitive hierarchy emerged on the basis of the clinical effort needed to implement the treatment: TCAs were generally used first, MAOIs second and ECT third for most patients. There were no randomized controlled trials (RCTs) of TRD in that era, though clinical wisdom taught that MAOIs worked in about one half of the cases in which TCAs failed, and ECT was expected to benefit at least 80% of antidepressant non-responders5. As a result, clinicians might have predicted that 70-90% of depressed patients could be treated effectively with this three-step proto-algorithm. By the mid-1990s, selective serotonin reuptake inhibitors (SSRIs) and several other newer-generation antidepressants had supplanted the TCAs as first-line treatments, expanding that intuitive algorithm to four levels of treatment. When we first reviewed the growing literature on this topic5, 6, we reached a similar conclusion: a four-step treatment algorithm, in theory, might be expected to yield up to a 90% cumulative response rate. Yet, the 1980s and 1990s ushered in an era of increasing methodological rigor, and the first RCTs of TRD began to emerge. Results indicated that our estimates were overly optimistic, and methodological conventions such as intention-to-treat analyses, which account for the impact of attrition on response rates, and use of "blinded" evaluators to minimize expectancy biases, revealed more sobering estimates of benefit. For example, in the STAR*D trial, the cumulative response to a sequence of four treatment trials was in the order of 50-60%. So, the public health problem of TRD turned out to be much larger than anticipated and, as reviewed by McIntyre et al, subsequent methods to refine and expand upon the simple hierarchical system that we had proposed have strengthened our ability to assess and classify depressions that do not respond to standard therapies. The introduction of a broader and more inclusive term, difficult-to-treat depression7, further enriches the conceptual framework of understanding the clinical context of non-response to antidepressants: there are many reasons that might explain why an antidepressant will not deliver the desired result, and only some of them pertain to neuropharmacological actions of our medications. Nearly 20 years have passed since the publication of the main findings of the STAR*D trial. In the post-STAR*D era, it can be argued that the greatest unmet need in the psychopharmacology of depression is for antidepressants that work via mechanisms other than modulation of monoaminergic neurotransmission. McIntyre et al provide a particularly useful summary of the data on several of the more recent therapeutic developments that have truly improved the outcomes of some people who do not respond to standard antidepressants. Switching antidepressants, which was once the quintessential second step in most algorithms, is now more of a default option for patients who have tolerability issues with the index antidepressant, and only rarely are patients switched to an MAOI. More commonly used second-line options include combinations of SSRIs and either mirtazapine or bupropion, and several adjunctive strategies. Among the adjunctive options, a large amount of empirical data supports use of a group of second-generation antipsychotics (SGAs). Given the well-known risk of weight gain, the potential for other metabolic side effects, and a small but real ultimate risk of tardive dyskinesia, more extensive data from longer-term studies are sorely needed to help to more accurately gauge the relative merits and cost-effectiveness of this adjunctive strategy. Although clinically tested and widely used, combining antidepressants and adjunctive therapy with SGAs can be thought of as incremental options, because they target somewhat complementary monoaminergic mechanisms and require that patients continue to take an SSRI or other newer antidepressant. There was a frustratingly long pause between the introduction of the various members of the so-called "newer" generation of antidepressants – it is, after all, more than 35 years since the US Food and Drug Administration (FDA) first approved fluoxetine – and the discovery of interventions with truly novel mechanisms of action. Fortunately things are changing, with the serendipitous observation that a sub-anesthetic intravenous dose of ketamine could have large and remarkably rapid antidepressant effects. Now confirmed by the findings of a large number of RCTs in patients with various forms of TRD8, the relatively rapid acceptance of this "off-label" use of intravenous ketamine has opened the gates to a new wave of potential therapies that target glutamatergic neurotransmission. While it remains to be seen whether intravenous ketamine or intranasal esketamine – the first FDA-approved therapy to result from these observations – will continue to be widely used a decade from now, it is a fact that the paradigm for drug discovery for TRD has changed for the foreseeable future. In this respect, the path for studying the therapeutic potential of neurosteroid drugs such as zuranolone, which is thought to indirectly affect glutamatergic neurotransmission through GABA-A receptor positive allosteric modulation, has been much less arduous than previously possible. Likewise, the paradigm change determined by the proven efficacy of intravenous ketamine, a controlled substance with abuse liability and characteristic dissociative effects, has prepared the field for a new wave of studies examining the therapeutic benefit of psilocybin and related psychedelic compounds that were once considered essentially off limits for therapeutic research. Finally, descendants of transcranial magnetic stimulation, including intermittent theta burst stimulation (iTBS) and an accelerated high-dose iTBS protocol utilizing magnetic resonance imaging to guide or target functional connectivity9, have given hope for the possibility of viable alternate neuromodulation strategies for patients with more advanced levels of TRD. McIntyre et al's outline of the evidence concerning TRD, therefore, is timely and provides a thought-provoking overview of an exciting new era in the therapeutics of depression.