James Murrough, MD, PhD

Ask the Experts
James Murrough, MD, PhD

Insights Into Treatment-Resistant Depression: Current and Emerging Therapies

Practice Community
New York, NY
Practice Niche
Hospital and Institute Affiliations
Icahn School of Medicine at Mount Sinai


Why are high rates of partial or inadequate responses to first-line antidepressant therapy seen in patients with major depressive disorder (MDD)?


If you think about how antidepressant medications were originally discovered, it was not based on a deep understanding of the nature or causes of depression, but rather it was an accidental observation. Certain classes of medications were used for completely different purposes; for example, tuberculosis drugs — astute clinicians noticed that when used in patients with comorbid depression, the patient’s mood lifted and they felt better. The modern era of first-line antidepressant treatments include selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs). These new medications or compounds functioned like the original drugs and were discovered through serendipity; the older drugs include the tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs).

I believe that the reason we do not have as high a success rate as we wished we did is because we are using tools that do not spring from a basic understanding of the biology of the disease we are trying to treat. It is not surprising that they are not as effective as we wish they would be.


What are the risks associated with partial or inadequate MDD treatment?


Depression is the number-one cause of medical disability in the world, partially because it is relatively common. Unlike many significant medical diseases like cardiovascular disease and diabetes that tend to affect older individuals, major depression tends to affect young individuals at a vulnerable time when they are establishing their identity, going to school, or trying to start relationships or families.

For some individuals, depression tends to run a chronic course involving periods of severe symptoms and periods of relative remission from symptoms. In a subgroup of individuals, it tends to recur and the longer the symptoms of depression last, the more havoc is brought into their lives in terms of being able to function. In more severe cases, individuals with depression may be at risk for suicide, which is a major public health problem in this country.

Despite improving rates for many medical conditions, such as cancer or cardiovascular disease, the outcomes from psychiatric diseases like depression have not improved in this country in the last 10 years and may have even worsened a bit. Individuals with untreated depression continue to be at risk for suicide, for major functional role limitations, and for general disability from symptoms.

There are data that show that the longer depression goes untreated, the more difficult it can be to treat. For example, if a patient is treated with a first-line antidepressant or psychotherapy while in a major depressive episode but they do not improve or only have a partial response, then their chance of having a recurrence of depression is greater. In general, we encourage clinicians and patients to aim to treat depression to a full remission of symptoms to avoid the risk for relapse, which can occur when patients have residual symptoms.


What are the current challenges associated with treating patients with selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs)?


First, I believe there is still significant stigma associated with treatment; understandably, patients do not want to take medication if they do not have to. There can be a negative reaction when prescribing medication, for example an SSRI, because depression is still something that is a little bit difficult for patients or even physicians in some cases to conceptualize as a genuine health problem that requires medical treatment. This is because, while depression is a medical disorder that often requires medication to relieve symptoms, you cannot necessarily readily observe outward manifestations of the disease. There is no blood test for depression; rather, it is diagnosed based on a pattern of symptoms and history. Diagnosing depression is different from diagnosing other medical conditions, where you can locate a part of the body that is injured and prescribe medication that will help. With depression it is more complicated; it affects your brain and your sense of self.

The way we work with patients is basically through education, by explaining that depression is a medical condition. There is a lot of research showing that depression is a brain disorder. There are systems in the brain that regulate mood, affect, and energy, and there are abnormalities in these systems that lead to symptoms of depression. We try to work with the patient to overcome the stigma or any feelings they may have for not wanting to take medicine. In general, we know that across all areas of medicine, many patients do not take their medications as prescribed; depression is no different. The estimates of patients who take their medications as prescribed are between 40% and 60%. This is in part because patients do not like the idea of taking a pill because they do not like the idea that they are sick, particularly sick with something that is hard to define like depression.

The second reason patients do not take their medicine is because of side effects. Some of the most common side effects are sexual side effects (eg, making it more difficult to have an orgasm), and these types of side effects can really affect individuals. One strategy is to discuss these side effects with the patient upfront and explain what to expect. This allows the patient to understand the side effects and decide whether it will be a deal breaker or whether they can live with the side effects. I educate my patients on the way antidepressants work because unfortunately you get the side effects upfront but the benefits come later.

Early side effects in addition to sexual side effects would be upset stomach and gastrointestinal disturbances such as diarrhea, nausea, or other changes. There are nonspecific side effects such as headaches, not feeling “quite right,” sedation, feeling tired, and sometimes feeling activated or having too much energy. If the patient is not prepared for these, then they may very well stop taking their medications. This is where psycho-education becomes very important. I discuss with my patients that they might feel a little tired, a little jittery, or their stomach might feel a little upset; as long as it is mild, they should hang in there because it will go away within the first week. It takes probably closer to 2 weeks or longer for the antidepressant effects to actually start working.


Can you discuss the potential therapeutic role of modulating pathways outside of the monoamine oxidase pathway in MDD? This might include opioidergic, ionotropic, and metabotropic glutamatergic receptor modulation.


Going back to what I previously noted, the medicines we have now were not necessarily derived from any specific insights into the cause of depression. A lot of the new research is now focused on trying to understand whether some of these signaling pathways like glutamate or gamma-aminobutyric acid (GABA) could be a way to treat depression. When you treat patients with monoamine-based drugs, there is a treatment delay before you see measurable improvements in mood during which we envision that there are some downstream signaling effects. We know that chemically as soon as these medications are taken, effects on the serotonin or norepinephrine receptors are happening very quickly; however, there is this delay before improvements in mood are noted. So it must be that there are some downstream changes that involve other systems that are not directly targeted by current antidepressants.

Systems that involve glutamate, GABA, and others I think hold substantial promise for new treatment discovery in that regard. There has been a lot of work on targeting glutamate inotropic receptors like the N-methyl-D-aspartate receptor, but there is no antidepressant currently on the market that affects any system other than these classic ones that we have been describing.

There is one research program has been looking at a combination of medicines that target opioid signaling that has shown promise in patients with major depression. The opioid system is interesting. On one hand we are in the middle of an opioid epidemic. Therefore, everyone is going to be very careful to make sure that we are not giving something to our patients that later we may find out is causing other problems like addiction or dependence. Nonetheless, data so far on some of these newer agents are looking promising in part because the opioid system is complicated; it has many different receptors and it seems that we have a pretty good sense of which receptors carry the abuse and dependence risk, like the μ-opioid receptor. There may be other receptors like the κ-opioid receptors that are involved more in mood and dysphoria. Blocking the κ-opioid receptors may cause antidepressant effects without experiencing euphoric or dependence-like effects through the μ system.

There has been a lot of excitement in the field. I would say the field is ready for therapies with novel mechanisms for depression, assuming that they are safe, well tolerated, and will improve on our current treatments. This is the focus of the field right now.


Can you briefly review your general treatment algorithms for patients with treatment-resistant depression?


When I evaluate a patient for depression, the first question I typically ask is do they have a major depression that needs treatment, and if so, are they better suited to medicine or psychotherapy? If a patient is having continued depression despite psychotherapy, I would add a medicine. If they are already on a medication and they still have ongoing depression, then the options are either to add psychotherapy, add a second medication, or switch to another medication.

If a patient reports they have depression and despite several different treatments have not yet responded, then we may describe this patient as having treatment-resistant depression. The first thing I would do is make sure — based on their history and review of pharmacy or medical records — that they have had an adequate trial of at least 2 or 3 first-line treatments.

It is important to keep in mind that for some patients, antidepressant effects can work as soon as within 2 weeks but for many it can take up to 4, 6, or even 8 weeks. When a patient reports that a treatment did not work, it is important to determine whether they had an adequate trial at a therapeutic dose for at least 4 to 6 weeks.

If they have had a trial of an SSRI and they do not respond, then I would try an SNRI like venlafaxine, desvenlafaxine, or duloxetine, or an atypical antidepressant such as mirtazapine or bupropion. Even within the class of SSRI medications, not all of the medications are created equal. Each of the first-line antidepressant agents — which includes all SSRIs, SNRIs, mirtazapine, and bupropion — is unique in its side-effect profile and one or another is likely the best fit for a particular patient. The challenge for the clinician (and patient) is that it can be a trial-and-error process to find the right treatment for the right patient.

Optimizing each medication trial is important, as is making sure that the dose titration is sufficiently slow so as to avoid side effects as much as possible. I usually start my patients on the lowest possible dose and then increase from there to minimize the possibility of having side effects. For patients who did not respond to 2 or 3 adequate monotherapy trials of these first-line agents, we start to think about either combining medications or an augmentation strategy.

One of the augmentation strategies with the most data would be adding a second-generation or atypical antipsychotic medicine to one of the first-line agents like SSRI or SNRI, typically at a dose a lot lower than is used to treat disorders like bipolar mania or schizophrenia. Several of these agents now have FDA approval for augmentation in major depression. For example, aripiprazole doses for schizophrenia can be as high as 30 mg a day or more but for depression we are looking more at 2 mg, 5 mg, or maybe up to 10 mg. One of these agents would be added to a first-line agent to boost the effect. Lithium has a lot of data for treatment of depression; again, the dose is usually lower than for bipolar I disorder. Lithium added to an antidepressant can be an effective strategy.

For combination therapy of more than 1 antidepressant, we often aim to use 2 medications from different classes because at least in theory we would be targeting different aspects of what might be going wrong in the brain. If a patient is on fluoxetine, we might add bupropion, which is a dopamine-norepinephrine-targeting drug or mirtazapine, because neither of these are SSRIs. In reality, available data comparing combination strategies with agents from the same or different classes do not clearly show that using agents from different classes is superior.

If there is no response after combination or augmentation therapy, which may include psychotherapy, always look for an occult problem that was missed. Is there a substance use disorder going on that we do not understand or a complicated medical condition? When you are not getting the response that you expect, you always want to go back and make sure the diagnosis is correct. Maybe the patient has a bipolar II disorder that does not respond well to antidepressants, and in that case, you would be thinking about medications like mood stabilizers.

TCAs and MAOIs are older classes of medications that we use that may have more of a side effect burden but can be very effective. We can then consider neuromodulation techniques, including transcranial stimulation and electroconvulsive therapy. Electroconvulsive therapy remains the gold standard for severe and refractory forms of depression.

Disclosures: Dr Murrough has provided consultation or scientific advisory services to Allergan, Novartis, Fortress Biotech, Janssen, Medavante-Prophase, and Global Medical Education.