Module 1: Advances in the Biology and Treatment of Depression

I would like to first review the learning objectives for the program. Once you have completed this program, you should be able to discuss the role of various neurotransmitters in the pathophysiology of depression, including norepinephrine, dopamine, and serotonin. You should be able to review the pharmacology of the MAOIs–tyramine-MAOI interactions and drug-drug interactions–and compare and contrast the pharmacokinetics of oral versus transdermal MAOI delivery. I would also like you to understand some current advances in the treatment of depression.

After completing this program, you should also be able to better understand the physician and patient barriers to appropriate therapy for depression, outline patient compliance issues that are related to the treatment of depression, and acknowledge effective interventions for improving compliance.

We will start with a discussion of the epidemiology and pathophysiology of MDD.

Epidemiology of Depression:

As many of you know, major depression is one of the most common of all the major medical disorders. Lifetime prevalence rates from the work of Ron Kessler and his colleagues at Harvard, who led the National Comorbidity Study, revealed that about 21% of women and about 13% of men will actually have a major depressive episode in their lifetime.

And, of course, one of the big worries we have as physicians is the concern about suicide. Last year, there were 28,500 suicides in the US. The vast majority of these occur in association with a psychiatric disorder, and, more specifically, with a mood disorder, such as major depression. There are a number of adverse consequences of untreated depression in addition to suicide. We also worry about medical comorbidity. It is not generally appreciated, for example, that the presence of major depression markedly increases the risk for patients to develop coronary artery disease. Indeed, the risk of major depression in contributing to the development of heart disease is as great as that of cigarette smoking.

In addition, if you have major depression and have an acute myocardial infarction, your chance of survival in the next 6 months is much, much less. And for that reason, it is very important for us to treat depression aggressively, to reduce the morbidity and mortality associated with it.

DSM IV: Symptoms of Major Depression:

To remind you about the diagnostic criteria, the DSM-IV criteria for major depression state that a person has to have five of these nine symptoms every day for a period of 2 weeks or longer, among them depressed mood, loss of interest in usual activities, loss of appetite with weight loss or an increase in appetite and weight gain. In addition, patients have a sleep disturbance—either insomnia or hypersomnia, which is oversleeping. We also see evidence of psychomotor agitation or retardation, decreased energy, feelings of worthlessness or excessive or inappropriate guilt, a terrible self-esteem, difficulty concentrating and making decisions, and, as I alluded to earlier, the concern about suicidality is, of course, paramount in our minds when treating patients with depression.

Major Depression Subtypes:

There are a number of depressive subtypes, and I will not be going into this in any great detail, but one subtype is particularly important here, and that is major depression with atypical features. These are patients that often show hypersomnia instead of insomnia, increased appetite, tremendous fatigue, and a symptom they describe as "leaden limb paralysis," in which they feel that their arms and legs are literally made of lead and that they can barely lift them. They also show an unusual degree of rejection sensitivity and personal disappointment, even in response to a mild setback. These patients often show reverse diurnal mood variation; that is, unlike most depressed patients, they actually feel better in the morning and worse as the day goes on.

It is generally estimated that about 20% or more of all of the depressed patients that we see fulfill criteria for major depression with atypical features. What causes depression? We know that this is a disease that is a prototype of a gene and environmental interaction. In order to develop MDD, one generally has to have some genetic risk, which accounts for about a third of the vulnerability for the disease, and one also has to have some environmental risk. This can take the form of stress, severe stress. Early-life trauma in the form of child abuse is a common example of the types of environmental stress that contribute to depression. Poverty is another. It can also take the form of a biological perturbation, such as inflammation, comorbid medical disorders, infection, and so on.

All of these environmental and biologically driven genetic contributions appear to act on the substrate within the brain of the neurotransmitter systems. As you can see here, norepinephrine, serotonin, and dopamine are the preeminent systems thought to play a major role, a preeminent role, in the biology of depression. In addition, there is a clear alteration in neuroendocrine function. This includes, in particular, the stress system. Many investigators conceptualize depression as a stress response gone awry. The critical biological player here is corticotrophin-releasing factor (CRF). This slide summarizes a number of perturbations in the CRF system and in the hypothalamic, pituitary adrenal axis that have been documented in depressed patients, both in studies in living patients and in postmortem studies of brain tissue of suicide victims. As you can see, the CRF system and the HPA axis, in general, are hyperactive in depressed patients. There are also other neurotransmittersystems likely involved in the biology of depression, including the GABA system, substance P, NPY, neuropeptide Y, and oxytocin.

Norepinephrine Alterations in Depression:

To summarize briefly, the norepinephrine system has been intensively studied. There are very clearly alterations in the concentration of the norepinephrine transporter or the NET as it is called, or the uptake site. This has been reported in postmortem studies of suicide victims, in the locus coeruleus, the site of origin of the norepinephrine cells that project 70% of norepinephrine to the forebrain. There are changes in both the alpha and the beta adrenergic receptors, receptors that norepinephrine acts upon within the brain. The up-regulation of the beta receptors is thought to be due to a relative deficiency of norepinephrine at certain critical synapses in the brain. In older studies, measurements of norepinephrine metabolites, such as 3-methoxy-4-hydroxyphenylglycol (MHPG), have been reported, as well.

Serotonin Alterations in Depression:

No system has been studied as much as the serotonin system. There is a large body of evidence that the serotonin system is awry in depression in many, if not most, patients. There is truly a real deficiency of serotonin in depressed patients. This has been shown in PET studies using ligands for various measures of the serotonin system, such as the serotonin transporter. It has been shown in blood platelets, which contain the serotonin transporter as well. It has been shown in postmortem tissue. Similarly, there have been changes reported in the density of the postsynaptic 5-HT2 receptors. Taking this together, one would suggest that the overwhelming evidence is of a relative deficiency of serotonin in the brains of patients with depression.

Dopamine Alterations in Depression:

In contradistinction to the norepinephrine and serotonin systems that have been scrutinized to a considerable extent regarding their potential roles in the biology of depression, the dopamine system has not really been explored in any considerable detail. This is somewhat surprising, because dopamine mediates our ability to experience pleasure from a variety of sources, whether it is eating great food, or sexual pleasure, or reward when we compete in various events.

We know that the cornerstone of major depression, symptomatically, is anhedonia, the inability to experience pleasure. So it would stand to reason that if this is, in fact, a major portion of depression, there ought to be alterations in the dopamine system. As you can see in this slide, there is increasing evidence that dopamine systems are pathologically involved in the biology of depression. Those findings include a reduction in the concentration of the major dopamine metabolite, homovanillic acid, in both the plasma and cerebrospinal fluid of depressed patients. A decrease in the density of the marker for the presynaptic dopamine neuron, the dopamine transporter, is shown in several PET studies, as well as an elegant PET study looking at an update of radiolabeled dopa, [18F] fluorodopa, and finally, a compensatory increase in the number of D₂ dopamine receptors in the striatum.

Areas of Increased Cerebral Blood Flow in Major Depressive Disorder:

There have been a number of other approaches to looking at the biology of depression. These would include functional imaging studies, not looking at a particular neurotransmitter constituent, but instead looking at the activity of different brain regions.

Just as a reminder, one can use either cerebral blood flow as a proxy of neural activity or glucose uptake; because as it turns out, different brain regions increase their blood flow and glucose uptake when they increase their activity, because brain neurons have a tremendous need for oxygen and fuel in the form of glucose. What you see here is an elegant study done by Wayne Drevets at the National Institute of Mental Health. Similar studies have been done by Helen Mayberg at Emory University. What you see here is increased activity in the amygdala. This is an almond-shaped organ in the temporal lobe of the brain that is an integral part of the limbic system, and we know this structure is indeed abnormal from imaging studies and postmortem studies in the brains of patients with depression.

SPECT Brain Imaging: 3D Rendering of Voxels Reflecting Lower Tracer Distribution in Depressed Patients:

When we have looked at other imaging modalities, such as SPECT, or single-proton- emission computed tomography, we can actually begin to compare different subtypes of depression. I know this slide is a little bewildering, and I am not going to spend a lot of time on it, but the important point is on the left of the graph, which shows differences between depressed patients and controls. The next two parts of the slide show differences between depressed patients and controls, but divide up the patients into those with atypical depression on the right and those with regular garden-variety, run-of-the-mill major depression in the middle of the slide. The important point is that biologically they look very different. Putting this together with some of the other data I have described to you, I would suggest that part of this may well be that atypical depression may have a larger role for dopamine in its pathophysiology.

That will, of course, have treatment implications, which I am going to discuss now in our section on management of depression.

Developmental Timeline of Depression Treatment Options:

This is a developmental timeline of depression treatment options, and they relate, literally, to the time that new treatments were introduced to manage this devastating disorder. We start with electroconvulsive therapy (ECT), as you can see here. After ECT, MAOIs were developed. These are the older, traditional MAOIs: phenylzine, phenylcipramine and isocarboxazid. These were developed after a few very astute clinicians noticed that patients treated with an MAOI for tuberculosis (TB) in sanitariums in Switzerland actually showed mood improvement, even when their pulmonary conditions did not improve. That led to the introduction of these agents to the depression pharmacopoeia along with, at about the same time, the introduction of tricyclic antidepressants (TCAs).

These medications, though effective, clearly had a number of drawbacks in terms of lethality and overdose, and a number of other troubling side effects both on the cardiovascular system. They also had some other troubling side effects that led to a decrease in patient adherence. That led to the development of newer agents that were not lethal in overdose and did not share those adverse properties. These include trazodone; the selective serotonin reuptake inhibitors (SSRIs), of which you know many are on the market; nefazodone; the cousin of trazodone, bupropion, an antidepressant whose mechanism of action remains relatively obscure, certainly not the reuptake inhibitor we thought it once was; and mirtrazapine, a drug that is not a reuptake inhibitor at all, but acts on receptors, serotonergic and noradrenergic. A little later, venlafaxine and then duloxetine, drugs that act by blocking both serotonin and norepinephrine, were introduced. Most recently, the selegiline transdermalsystem was approved, a method to administer a MAOI with a much more favorable side-effect profile.

Obviously, there are a number of nonpharmacologic treatments, both older and newer. They include ECT, which of course was first developed in the 1940s for the treatment of schizophrenia and later found to be very effective for the treatment of depression. More recently, vagal nerve stimulation (VNS) and deep brain stimulation (DBS) were approved by the US Food and Drug Administration (FDA) for the treatment of refractory depression. An experimental treatment, repetitive trans-cranial magnetic stimulation (rTMS), is in clinical trials. And of course, as you will see, psychotherapies were developed to treat depression.

Guidelines for Treatment of Major Depression:

Slide 13.

This slide summarizes the American Psychiatric Association guidelines for the treatment of major depression. I know many of you are familiar with these guidelines. They are, of course, an important reference. These guidelines were developed about 6 years ago, and they certainly will be revised relatively soon.

Antidepressant Pharmacotherapy:

Slide 14.

I would like to briefly review the current pharmacopoeia for the treatment of depression. This is a list of the currently available MAOIs. You see that there are two traditional MAOIs available, phenylzine and tranylcypromine. Isocarboxazid is not listed here because it is available in some states and not others. For that reason, many individuals who have tried to prescribe it have had difficulty doing so. Selegiline is certainly available, but in an oral form. I personally would not recommend using it for the treatment of depression for reasons that will come later. It is the selegiline transdermal system (STS) that one would want to use for the treatment of depression.

A variety of tricyclic antidepressants are listed here, as well as the one tetracyclic, maprotiline. These drugs, in my mind, should never be prescribed as first-line agents. Lethality in overdose is well known. These remain the number-one cause of drug overdose in the US, second only to heroin, number one in terms of prescribed medications, and certainly we do not want to give medications like this to patients who are depressed and might have an increased risk of suicide. They should be reserved for very refractory patients who have responded to no other, safer treatments.

In the SSRI category, remember that fluvoxamine is not approved for the treatment of depression in the US. It is an effective antidepressant, but it was never approved by the FDA because data were never provided to the FDA to support its approval. The other five all are effective, although I will remind you that in the recent STAR*D study led by John Rush of the University of Texas and funded by the National Institutes of Health, the remission rate associated with citalopram treatment was a disappointing 28%. So these medications, though they are clearly an improvement over the TCAs, are not the panacea for treatment of depression.

We then move on to the dual reuptake inhibitors, that is the serotonin and norepinephrine reuptake inhibitors, venlafaxine and duloxetine. I have recently completed a meta-analysis of all of the venlafaxine studies, comparing them with SSRIs. Although there was a slight statistical advantage compared with the SSRIs, whether this was clinically meaningful or not remains a debatable point. Again, clearly there is not as high a remission rate as we would have liked.

Finally, a number of antidepressants are grouped together because they are so-called atypicals. Nefazodone and trazodone are weak reuptake inhibitors that are primarily 5-HT2 receptor antagonists. Mirtazapine is an alpha-adrenergic and serotonergic receptor antagonist, and bupropion is a weak dopamine reuptake inhibitor. At best, these are antidepressants whose mechanisms of action are not clear.

Measurement-Based Care and Outcomes With Citalopram for Depression: STAR*D Report:

This is the STAR*D data that I was alluding to earlier, showing remission and response rates over time in a large number of patients—1,343 in the response category, 943 in the remission category. As you can see, the results are quite disappointing: over a 13- to 14-week period, we see remission rates of 28% using the HAM-D and 33% using another dimensional rating scale, the QIDS-SR. Response rates, defined as a 50% or greater improvement in symptom severity, were less than 50%. All of us would agree that this is quite problematic and suboptimal.

Efficacy of Selegiline in Depression:

Slide 16.

These are the data on an oral selegiline study done by John Mann, a professor at Columbia University in New York. You can see here that the first 3 weeks of treatment involved 10 milligrams per day. There was no statistically significant difference with placebo in terms of the Hamilton Depression Rating Scale score. When the dose was increased an average of 30 milligrams a day (some patients had higher doses), you see a clear antidepressant effect over the 6-week treatment period.

Efficacy of Common Antidepressants for Continuation Treatment of Major Depression:

Slide 17.

When you consider the results that I have discussed with you thus far, and now consider the kind of meta-analysis that is often performed with large databases such as this one that was reported in the Lancet, the take-home message is very clear. Although our current antidepressant treatment is clearly better than placebo, it is not optimal. Also, we clearly see higher relapse rates with placebo than we do with active treatment. But the bottom line is that in looking at long-term treatment, we simply do not see the kind of rates in keeping people well and preventing relapse that we would all consider minimally acceptable. Part of that is a tolerability issue. And partly the problem is that, unlike other branches of medicine, it is very hard for us to match a given patient to an effective treatment. Put another way, we have no predictors of response to one or another antidepressant.

Psychotherapy:

Slide 18.

I would be remiss if I did not discuss psychotherapy. It is my own personal view, based on the data, that psychotherapy plus antidepressant treatment is more effective than either treatment alone. Our studies done with Marty Keller at Brown University looking at chronic depression have certainly supported that point of view. In addition, I have done a number of studies myself in recent years that demonstrate that, particularly in individuals with early life trauma such as child abuse, psychotherapy alone is more effective than pharmacotherapy alone. I still believe the combination is most effective. This slide summarizes the effectiveness of interpersonal psychotherapy on the left and cognitive behavior therapy (CBT) on right. I implore all of you to know who are the psychologists or psychiatrists or social workers trained to deliver interpersonal psychotherapy (IPT) or CBT in your community, and refer patients to those practitioners. This will not only enhance patients' treatmentresponse, it will also enhance compliance with the medications you prescribe.

Cognitive Behavioral Therapy (CBT): Antidepressant, CBT, or Combination Therapy in Major Depressive Disorder:

This is from that study that Dr. Keller and I and a number of others were involved in, showing that combination therapy with the antidepressant nefazodone plus CBT is more effective in treating chronic depression than either CBT or the antidepressant alone. The CBT used was a modified form of CBT that was developed by Jim McCulloch at Richmond. You will also notice that those two modalities had equal efficacy in this chronic depression population.

Current and Future Nonpharmacologic, Nonpsychotherapeutic Treatment Options:

Slide 20.

Let us talk for a few minutes about nonpsychotherapeutic, nonpharmacological treatment options. ECT is very effective, no question about it. But it is only done at a limited number of centers. The problem with ECT, in my mind, is that once you have responded, unless you are enrolled in a maintenance ECT program you are going to begin losing the gains that you have obtained with ECT at about a rate of 4% per day. That is obviously a serious problem. Additional treatments in this category are VNS, approved by the FDA for treatment of refractory depression, and then three novel somatic treatments: rTMS; magnetic seizure therapy, which uses a magnetic field to actually induce a seizure locally within the brain; and DBS, the newest of treatments, studied by Helen Mayberg and her colleagues at Emory University.

Acute- and Long-Term VNS Efficacy in the Treatment of Major Depression:

These are the data that led, at least in part, to the approval by the FDA of VNS for the treatment of refractory depression. It compares treatment-as-usual with treatment-as-usual plus VNS. As you can see here, there is an advantage of VNS added to treatment-as-usual. Many patients in this study had already been treated with combinations of antidepressants and/or psychotherapy.

ECT vs rTMS in Treatment of Major Depression:

These are data from Phil Janicak at the University of Illinois. This study, comparing rTMS with ECT, showed similar efficacy of these two modalities. If this ends up being true, rTMS, which is much easier to administer, certainly will be adopted and approved by the FDA for the treatment of depression.

Repetitive Transcranial Magnetic Stimulation:

These data are from Holly Lisanby at Columbia University in New York, who looked at Hamilton Depression Rating Scale scores pre-rTMS and post-rTMS, and again you can see the impressive antidepressant effect of this modality.

Summary of rTMS Clinical Trials in Major Depression:

This table summarizes rTMS clinical trials and includes data that were presented at the American Psychiatric Association meeting in New York and then data presented in Toronto in 2006. As you can see, there is increasing evidence that rTMS is effective in the treatment of depression compared with either sham treatment or those open studies that were conducted showing efficacy as well.

Deep-Brain Stimulation:

Finally, Dr. Mayberg conducted a study in Toronto of 6 extremely treatment-refractory patients. These are patients who failed multiple medication trials, ECT, and so on. An electrode was placed to stimulate the anterior cingulate. And a remarkable antidepressant effect was observed in 4 of the 6 patients studied. Clearly, we need to do a lot more work in this area in order to understand how deep-brain stimulation works in treating refractory depression.

Treatment-Resistant Depression:

Speaking of treatment-resistant depression, I thought it would be best if we went ahead and actually defined this diagnostic entity. The time-honored definition of treatment-resistant depression is a patient who has failed two adequate trials with antidepressants from disparate pharmacologic classes; for example, a patient who has failed in a trial with citalopram and then went on to fail in a trial with mirtazapine.

The prevalence of this entity is variously defined, depending on whether you are talking about total treatment resistance or partial treatment resistance; but it is thought to be somewhere between 10% and 30%. I think it is actually somewhat higher, more in the 40% range, if you include patients who do not attain remission. Some people think it may be in the 50% to 60% range. When thinking about patients who are treatment failures, as they are called, one ought to think about the potential reasons for failure.

One reason for treatment failure is lack of adherence to the prescribed medication—many patients simply do not adhere to their prescribed regimen. They may miss doses or simply not be taking their medication at all. This is a good reason to check a blood level of any antidepressant. Patients may have a comorbid and occult medical disorder, such as hypothyroidism or anemia or a brain tumor or epilepsy. They may be taking medications that are precipitating depression, such as oral contraceptives or glucocorticoids, steroids that are used to treat asthma or rheumatoid arthritis. They may have an occult psychiatric disorder that contributes to treatment refractoriness, such as occult alcoholism or substance abuse. And finally, the patient may have a subtype of depression, such as atypical depression or psychotic depression, that simply has not been addressed in the development of a treatment plan.

Managing Treatment-Resistant Depression:

How do we manage treatment-resistant depression? What is the current thinking about this? In general, the first thing we consider is optimizing the current therapy, making sure the patient is taking the medication, and then optimizing the dose. Although there are not a great deal of data about this, most practitioners, myself included, believe that if a patient is taking an SSRI, for example, and feels a little bit better but not much better, probably the best approach is to increase the dose, particularly if the patient has no side effects. There will be a certain percentage of patients who are very rapid metabolizers, who will require super-normal doses of antidepressants.

A second approach is combination therapy, to combine antidepressants from two different classes. This can, of course, have the propensity to increase the side-effect burden, but it certainly has been shown to lead to an improved response. For example, combining mirtazapine with SSRIs leads to an improved response compared with treatment with an SSRI alone. Another approach, which I have already alluded to is the combination of pharmacotherapy and psychotherapy. Finally, there is augmentation, the notion that frankly there are medications that in and of themselves are not antidepressants, such as thyroid hormone, triode, or thyronine or T-3, or atypical antipsychotics or dopamine agonists like pramipexole or ropinirole or lithium. These are all agents that, by themselves, are not effective in the treatment of unipolar major depression. However, when added to an antidepressant, they often convert nonresponders to responders.

Remission Rates With Lithium or T₃ Augmentation Following Two Failed Antidepressant Trials: STAR*D Report:

This is an example of such an approach. These are patients that had failed two antidepressant trials in the STAR*D study and were treated with either lithium or T₃ as augmenting agents. There was no statistical difference in their response, but T₃ certainly had a numerical advantage. More important, T₃ had much less of a side-effect burden than did lithium. One of the favorable reports from STAR*D is that every time another strategy was used in refractory patients, some patients did, in fact, convert to remitters. That is exactly what we need to know in developing our new algorithms for refractory depression.

Treatment Response Following 3 Failed Antidepressant Trails: STAR*D Report:

This next slide shows a very severely refractory group of patients. These are patients who failed three antidepressant trials. You can see that the numbers were small, about 45 to 50 patients per group. Having failed three trials, these patients were randomized to a traditional MAOI, tranylcypromine, or a combination of venlafaxine plus mirtazapine. As you can see here, they both showed an increased effectiveness over time at 14 weeks. The combination of venlafaxine and mirtazapine appeared to work faster than the classic MAOI.

Depression With Atypical Features:

Let me focus a little bit more on depression with atypical features. I have already mentioned that this is a subtype of depression that probably captures about 20%, in general, of the entire major depression population. Atypical depression is characterized by hypersomnia, that is, over-sleeping. Patients will talk about staying in bed for 16, 18 hours a day, sleeping for 12 to 14 hours, having leaden limb paralysis, being extremely rejection sensitive, and exhibiting an increase in appetite, and an increase in body weight. It does occur more commonly in women than in men and has been suggested to have an earlier age of onset than garden-variety, run-of-the-mill major depression.

Atypical Depression: Treatment Options:

The treatment options for atypical depression have traditionally focused on MAOIs. The late Fred Quitkin at Columbia and his group there, with Jonathan Stewart now one of the leaders, has clearly shown that MAOIs are particularly effective in atypical depression. That group, and Michael Thase's group at Pittsburgh, have repeatedly shown that many patients with atypical depression who fail conventional therapies do very well indeed with MAOIs.

However, SSRIs have been studied as well. In a study by Pat McGrath, Fred Quitken, and Jonathan Stewart, 150 patients reported a low placebo response rate and equal efficacy between fluoxetine and imipramine. A second study, by Strada and colleagues, showed similar results. In general, in most studies, except the one by Pande, MAOIs were actually superior to SSRIs.

TCA Treatment in Depressed Patients With Atypical Features:

This is one of the studies by Jonathan Stewart. The take-home message from this very busy slide is that early-onset patients with atypical depression had a poorer response to TCAs than they did to MAOIs. So think about early-onset, which is defined in that study as less than 20 years of age.

Efficacy of SSRIs in the Treatment of Atypical Depression:

In terms of the subsequent data, you can see here a study comparing placebo, fluoxetine, the prototype SSRI, and imipramine. In this particular study of about 50 patients, both fluoxetine and imipramine were superior to placebo, but the outcome was not great in either case.

Efficacy of Nonselective MAOIs in Atypical Depression:

The Pande study, which I believe was done when Pande was at Lilly, compared 20 patients with atypical depression treated with fluoxetine and 20 treated with the nonselective irreversible MAOI, phenelzine. Efficacy was about equal. Certainly other data and my own experience would suggest that MAOIs are superior to SSRIs and TCAs. It is surprising how little we really know about SSRIs in atypical depression. Very little data exist for S-citalopram, sertraline, and paroxetine.

Efficacy of Selegiline in Atypical Depression:

This is a study that the Columbia group did with oral selegiline. You will remember the Mann study I showed you earlier; this study used a similar regimen, except patients received 10 milligrams a day for the first 4 weeks, 20 milligrams in week 5, 30 milligrams in week 6. Clearly, compared with placebo, selegiline is a superior treatment.

And now it is my pleasure to turn the program over to Dr. Lindsay DeVane, who will review for you monoamine oxidase inhibition.