Jonathan E. Alpert, MD, PhD

Harnessing data from EMBARC, the largest placebo-controlled depression trial involving different neuroimaging approaches, investigators used machine learning to derive a unique electroencephalographic pattern or “signature” that appears to predict response to antidepressants vs. placebo. The methodological approach they used to further validate their finding sets an important example for future studies on EEG biomarkers with potential clinical relevance.

Randomized, placebo-controlled antidepressant trials have repeatedly demonstrated the modest superiority of antidepressant over placebo in the treatment of major depressive disorder (MDD).

 Correspondingly, only 1 in 3 depressed individuals remitted on a first trial of antidepressant in the classic STAR*D trial despite measurement-based care and active dose titration.

 These sobering and well-known realities appear to be less related to the weakness of available antidepressants than to the heterogeneity of depression; some individuals respond robustly while others remain depressed.

Roland Kuhn, a pioneer in the development of tricyclic antidepressants, 

the heterogeneity succinctly over 60 years ago.

 After describing exuberant successes on imipramine, he wrote:, “In many cases, however, there is merely some degree of improvement, making the condition more bearable for the patient, and even permitting the resumption of work, though at the cost of considerable effort. In other cases, there is no effect at all.”

Given the pervasive disability and suffering associated with depression, the long duration of antidepressant trials in the individual patient, and the uncertainty of the ultimate outcome, the search for meaningful predictors of response has been crucial but unfortunately painstaking and, with the possible exception of pharmacogenomics markers, largely unyielding to date of biological predictors ready for clinical adoption. Amidst the cacophony of biomarker studies, 

(EEG) signature putatively associated with predicting response to antidepressant vs. placebo.

 have pointed out, among various potential biomarkers, EEG, in contrast to peripheral body fluid measures, has the appeal of direct relationship to brain activity while also being more cost-effective and readily implemented in a variety of clinical settings than functional magnetic resonance imaging (fMRI) and without the radiation exposure incurred in nuclear medicine scans (PET/SPECT). Nevertheless, the literature on EEG and prediction of antidepressant response is largely characterized by publication bias toward studies with strong apparent positive effects together with lack of replication or validation in other samples.

In addition, EEG findings predictive of response in a study sample were associated with nonspecific prediction of response to any treatment or placebo rather than prediction of response to a specific treatment, thereby of questionable relevance to clinical practice in which key questions relate to selection among different agents and treatment modalities. In addition, few EEG studies have compared prediction models based on EEG with prediction derived from other kinds of brain related data such as fMRI or PET resulting in a biomarkers literature which is quite siloed and difficult to integrate.

While much work remains, the intriguing and rigorous study by Wu and colleagues

 suggests a blueprint of progress on all these fronts. In this research, the investigators designed a machine-learning algorithm for resting-state EEG (rsEEG) which they named SparseEEG Latent SpacE Regression (SELSER). They then applied the SELSER algorthim to clinical outcome data derived from the Establishing Moderators and Biosignatures of Antidepressant Response for Clinical Care for Depression (EMBARC) Study, the largest (n=309) placebo-controlled randomized clinical study of antidepressant efficacy that included neuroimaging markers including EEG and fMRI across four study sites using somewhat different EEG set-ups.

Using this machine-learning approach and honing their predictive models to find one that best fit the data across the four sites, they were able to discern an rsEEG signature or pattern at baseline which appeared to be uniquely associated with subsequent response to the selective serotonin reuptake inhibitor (SSRI) antidepressant prescribed in the study (sertraline) while not being associated with response to placebo. That is, the signature seemed to be a marker not for positive outcome generally but to an antidepressant specifically. They then validated the signature in an unrelated sample of antidepressant treated patients in a naturalistic, longitudinal depression study involving various antidepressants and found that the rsEEG signature was both related to antidepressant response, as in EMBARC, and negatively correlated with the number of failed antidepressant trials as would be expected from a biomarker associated with positive antidepressant outcome.

Another question the investigators addressed had to do with whether the rsEEG signature was related to another putative predictive signature that was derived from data in the EMBARC study, one related to fMRI conducted during a so-called emotional conflict task in which participants were presented with emotionally congruent stimuli (such as a fearful face with the word “fear”) or incongruent stimuli (such as a fearful face with the word “happy”) while in the fMRI scanner. They found significant convergence between the rsEEG- and the task-FMRI-derived machine learning predictions, a result that suggests that these two predictive signatures, though derived from different brain imaging modalities, may relate to some common neurobiological phenotype that is associated with responsivity to antidepressants.

Finally, the investigators wanted to know whether the rsEEG signature that predicted response to antidepressant treatment and not to placebo also predicted response to other active treatments. In order to address this question, they studied whether the rsEEG signature predicted response in an independent MDD treatment study of repetitive transcranial magnetic stimulation (rTMS) of the dorsolateral prefrontal cortex (DLPFC) and concurrent psychotherapy.

Rather than finding a convergent relationship in which the rsEEG signature that had predicted antidepressant response in EMBARC also predicted response to rTMS and psychotherapy, however, they found an opposite relationship. Indeed, individuals who had a positive response to rTMS and psychotherapy proved more likely to have an rsEEG signature associated with poor response to sertraline. This finding further supports the potential specificity of the rsEEG signature; it is neither associated with positive response to placebo nor to rTMS and psychotherapy but appears to be associated with positive response to antidepressants per se.

Among the ways such a signature, if confirmed, could advance MDD treatment is that it might help clinicians to identify patients who are good candidates for assertive pharmacotherapy vs. those who might benefit more by fast-tracking to other treatments such as psychotherapy or neuromodulation. Such a signature could help reduce the high rates of placebo responses in clinical trials by pre-selecting subjects with the greatest likelihood of responding to pharmacological agents vs. placebo. Finally, when combined with other probes of brain function, it may help further define the neurobiological phenotype associated with responsiveness to antidepressants generally or particular antidepressants.

In order to more fully understand and trust the potential clinical application of this EEG biomarker for predicting response to antidepressants, additional research is clearly needed. At the very least, this work needs to be extended to diverse, real-world clinical samples receiving a broader range of antidepressants as well as to individuals receiving TMS, ECT, and different forms of evidence-based psychotherapy. It also needs to be compared with findings from other brain imaging techniques beyond the specific task-related fMRI measure in EMBARC.

Nevertheless, the study by Wu and colleagues

 has helped elevate the standard by which other EEG biomarker studies can be assessed; namely they replicated their proposed rsEEG signature in independent samples, found evidence for convergence with other brain biomarkers (fMRI), and were able to distinguish prediction of response to antidepressants from prediction of response to other forms of active treatment or to placebo. These are vital steps on the long but much needed journey toward personalized medicine approaches in depression treatment.

is the Dorothy and Marty Silverman Chair of the Department of Psychiatry and Behavioral Sciences at Montefiore Medical Center and Albert Einstein College of Medicine where he is Professor of Psychiatry, Neuroscience, and Pediatrics. The author reports that he receives research support from Otsuka Pharmaceuticals, Axsome Therapeutics, and the Patient-Centered Outcomes Research Institute. 

1. Cipriani A, Furukawa TA, Salanti G, et al. Comparative efficacy and acceptability of 21 antidepressant drugs for the acute treatment of adults with major depressive disorder: a systematic review and network meta-analysis. 

2. 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. 

The treatment of depressive states with G 22355 (imipramine hydrochloride)

An electroencephalographic signature predicts antidepressant response in major depression

5. Widge AK, Bilge MT, Montana R, et al. 

Electroencephalographic biomarkers for treatment response prediction in major depressive illness: a meta-analysis

Articles

Investigators used machine learning to derive a unique electroencephalographic pattern or “signature” that appears to predict response to antidepressants vs. placebo.

Homing in on an EEG Signature to Predict Antidepressant Response 

Second-generation antidepressants have served as first-line treatments for major depressive disorder (MDD) for over two decades. Nevertheless, the nature of the dose-response curve for contemporary antidepressants has proved surprisingly difficult to define.

comparing the efficacy of fixed doses described a flat dose-response curve, a conclusion that seemed well aligned with receptor binding analyzed by positron emission tomography (PET) in which 80% serotonin transporter occupancy in the striatum appeared to be achieved at minimal therapeutic doses for most SSRIs and venlafaxine with only small incremental increases occurring at higher doses.

that used both fixed and flexible dosing strategies have somewhat modified assertions of a flat dose-response curve, suggesting modestly greater efficacy for higher doses yet still concluding that efficacy may plateau in the moderate dose range at around the equivalent of 40 mg to 50 mg of fluoxetine.

In contrast, standard clinical practice has been to increase doses more assertively as tolerated, particularly in the setting of 

 an approach that is also in line with the American Psychiatric Association (APA) guidelines for MDD treatment.

 conducted the most sophisticated and comprehensive meta-analysis to date of fixed-dose, double-blind, randomized controlled trials of second-generation antidepressants, both published and unpublished, for the acute treatment of MDD.

Rather than limiting their analysis to efficacy (as assessed by 50% reduction in depression severity), the investigators also assessed treatment tolerability (as assessed by dropouts due to adverse effects) and treatment acceptability (as assessed by dropouts for any reason). In all the researchers analyzed 77 studies involving 19,364 participants. The mean age of participants was 42.5 years and 60.9% were female. The median length of antidepressant trial was 8 weeks. They conducted separate dose-response analyses for individual SSRIs as well as for venlafaxine and mirtazapine.

For SSRIs they found that the dose-response curve for efficacy showed a gradual increase in doses of 20 mg to 40 mg fluoxetine equivalents, with either asymptotic or decreasing efficacy at higher doses. At the same time, the dose-response curve for drop-outs due to adverse effects increased linearly across all doses studied. Overall, the relationship between dose and acceptability overall favored the 20 mg to 40 mg fluoxetine equivalents. The dose-response curve for efficacy for mirtazapine appeared to peak around 30 mg and then decline. In contrast, the dose-response curve for efficacy of venlafaxine, though most sharply increasing between 75 mg and 150 mg, continued to increase steadily up through 400 mg, the highest dose studied.

Nevertheless, dropouts due to adverse effects or other reasons increased with dose for both mirtazapine and venlafaxine indicating greater acceptability at low to moderate doses (around < 40 mg and < 200 mg respectively). Based on their overall findings, the authors conclude, “For the majority of patients receiving an SSRI, venlafaxine, and mirtazapine for the acute-phase treatment of their major depressive episode, the lower range of the licensed dose will probably achieve the optimal balance between efficacy, tolerability, and acceptability. Clinical guidelines need to incorporate these findings.”

Where does this leave the clinician who values the evidence base yet recognizes the risks and burden of unremitted depression and not infrequently advocates for patients in their appeals to insurers reluctant to cover daily antidepressant doses above the usual range?

On balance, the study provides further evidence that for some patients there may be diminishing returns, and, indeed, potentially lower response rates at SSRI doses that exceed 20 mg to 40 mg of fluoxetine equivalents or at mirtazapine doses above 30 mg. For venlafaxine, the potential for response may continue to increase with dose escalation but so will the potential for dropout due to adverse effects. Nevertheless, this conclusion may well extend to only a subset of patients seen in psychiatric settings.

 found that of 599 patients with unipolar nonpsychotic MDD, a remarkable 79.5% would have been excluded from RCTs by usual eligibility criteria such as medical or psychiatric comorbidity and severity. In this context, Furukawa and colleagues

 explicitly excluded studies of MDD in patients with medical comorbidity. Moreover, many RCTs included in their analysis excluded individuals with significant psychiatric comorbidity such as substance use disorders, posttraumatic stress disorder, or severe personality disorders with active suicidality, or with a history of non-response to antidepressants.

In addition, fixed-dose studies do not mimic usual clinical practice. Clinicians do not routinely start treatment with moderate to high doses of antidepressants in the way patients in fixed-dose studies are directly randomized to these higher doses vs. lower doses. Rather, those patients in clinical practice who get to higher doses are typically those who have already proven themselves to tolerate lower doses and yet to have not remitted. The question at that point is, given reasonable tolerability and incomplete response, what is the probability of tolerability and greater response at a higher dose? Patients without adverse effects or benefit at a lower dose may be individuals who are rapid metabolizers of the medication or whose receptors are less responsive to usual doses.

Furthermore, for individuals who may have tried and failed other antidepressants, an additional question is whether the profile of risks and benefits involved in exploring higher doses is more or less favorable than starting from scratch with a new agent, augmenting the initial antidepressant with other agents, or combining it with second antidepressants. The Furukowa study

 was not designed to address these questions. Finally, although efficacy and tolerability during acute treatment are key factors, in clinical practice, efficacy and tolerability during continuation and maintenance treatment for prevention of relapse and recurrence are also integral to the overall impact and success of depression treatment.

The take home message is that, based on the most comprehensive and state-of-the-art meta-analysis of studies involving patients with MDD randomized to fixed doses of SSRIs mirtazapine or venlafaxine, the combined evaluation of efficacy, tolerability and acceptability, favors dosing in the low to moderate range in the simplest cases. Yet, as the evidence base does not extend well to the many patients with depressive symptoms seen in psychiatric practice with psychiatric or other medical comorbidity, before treatment resistance, or incomplete response but tolerability on a current antidepressant, the elegant meta-analysis by Furakawa

 does not provide sufficient reason to hold back on thoughtful dose increases as tolerated. Along with potential dose increases, in the setting of ongoing depression, the reassessment of diagnosis, comorbidity, and psychosocial factors together with consideration of augmentation and combination strategies, psychotherapy, and neuromodulation are also crucially important.

1. Meyer JH, Wilson AA, Sagrati S, et al. Serotonin transporter occupancy of five selective serotonin reuptake inhibitors at different doses: an [11C]DASB positron emission tomography study. 

2. Jakubovski E, Varigonda AL, Freemantle N, et al. Systematic review and meta-analysis: dose-response relationship of selective serotonin reuptake inhibitors in major depressive disorder. 

3. Ionescu DF, Rosenbaum JF, Alpert JE: Pharmacological approaches to the challenges of treatment-resistant depression. 

4. American Psychiatric Association. Practice guideline for the treatment of patients with major depressive disorder, third edition. 

5. Furukawa TA, Cipriani A, Cowen PJ, et al. Optimal dose of selective serotonin reuptake inhibitors, venlafaxine, and mirtazapine in major depression: a systematic review and dose-response meta-analysis. 

6. Zimmerman M, Chelminski I, Posternak MA. Generalizability of antidepressant efficacy trials: differences between depressed psychiatric outpatients who would or would not qualify for an efficacy trial. 

Although the current generation of antidepressants extends back to the late 1980s, efforts to describe the dose-response relationship for treatment of major depressive disorder continue. More in this research update.

Defining the Antidepressant Dose-Response Curve

With their early age of onset, high prevalence, chronicity, and pervasive impact on multiple domains of functioning, the burden of mood disorders exceeds that of virtually all medical conditions in the US and globally in terms of disability, cost, and suffering.

My long-term clinical and research interest in mood disorders has two primary sources. First, having a parent with a recurrent mood disorder taught me that vulnerability to depression can freely coexist with notable resilience and strength in many other areas, and that depression can be treated. Second, as a psychiatrist-in-training, I learned that mood disorders may be treatable for some but are barely treatable for far too many others. The field continues to evolve, but the challenges I first learned as a trainee largely remain.

Nevertheless, we have seen major advances over the past decade, including:

• Implementation of routine screening for depression in primary care and some sub-specialty care settings;

• Increasing adoption of measurement-based care;

• Greater emphasis on chronic disease management approaches for those with persistent or recurrent illness;

• A growing range of evidence-based psychotherapies and neuromodulation strategies;

• The quite preliminary yet promising emergence of combinatorial pharmacogenomics and other biomarkers that may augment clinical judgment;

• The FDA approval in 2019 of two rapidly acting antidepressants leveraging novel neural mechanisms; and

• An expanding range of big data approaches that promise to extend the reach of patient-oriented science well beyond traditional clinical trials.

The increased interest in the health and treatment of individuals traditionally under-represented in mood disorders research-including ethnically, culturally, and racially diverse individuals and individuals from sexual and gender minority groups-is also quite encouraging.

As we celebrate progress, we also appreciate the magnitude of the remaining obstacles. The considerable heterogeneity across patients in presentation, course, and treatment responsiveness continues to beg the question of whether the current nosology of mood disorders is meaningful. It remains quite possible that major depressive disorder is a non-specific manifestation of multiple pathophysiological entities. Along with high placebo response rates in depression studies, the potential inclusion of individuals whose symptoms reflect quite distinct etiologies means that the signal of promising treatments is almost certainly often lost amidst the noise of multiple subgroups with diverse responses.

Special patient populations with mood disorders, such as pediatric, geriatric, or peripartum, have been remarkably under-studied. And, access to expert care for individuals in these groups is often limited. As we welcome the burgeoning number of treatments and treatment modalities, we lack adequate research devoted to their optimal combination and sequencing.

In this Special Report, we have tried to target these important challenges with articles on the integration of psychotherapy and pharmacotherapy in the treatment of major depressive disorder and on diagnostic and treatment challenges in pediatric bipolar disorder.

The profound impact of mood disorders in health care settings and on society is widely acknowledged by clinicians, administrators, and policy makers. Government and health care leaders increasingly understand the value of greater investment in mood disorders. With increasingly sophisticated research approaches and evidence informed treatments, we have reason to look forward to meaningfully improved outcomes for the hundreds of millions of individuals who suffer from mood disorders during their lifetime.

Dr Alpert is Dorothy and Marty Silverman Chair, Department of Psychiatry and Behavioral Sciences, and Professor of Psychiatry, Neuroscience and Pediatrics, Montefiore Medical Center and Albert Einstein College of Medicine, New York. He reports no conflicts of interest concerning the subject matter of this Special Report. 

Author | Jonathan E. Alpert, MD, PhD

Articles

New content coming soon, please check back later.