What’s new in antidepressant treatment? Not much. Some existing antidepressants have received new FDA indications. For example, duloxetine (Cymbalta) was approved for generalized anxiety disorder in 2009 and for chronic musculoskeletal pain in 2010. Also in 2009, quetiapine extended release (Seroquel XR) was approved as an adjunctive treatment for major depression, joining aripiprazole (Abilify), which was approved for this indication in 2007. Other recent approvals have included escitalopram (Lexapro) for depression in adolescents, and Symbyax, the combo of olanzapine and fluoxetine, for treatment resistant depression.
In terms of “new” medications, in this article we will examine four recent additions. Two are essentially old wines in new bottles (Oleptro and Silenor), one is a reissue of an MAOI (Marplan), and one is actually a new molecule (Viibryd). You may have heard some buzz about agomelatine as well, a unique melatonin agonist, but it is too far away from FDA approval (probably 2012 at the earliest) for us to cover it here.
Vilazodone (Viibryd) was approved by the FDA in January of 2011, and should hit pharmacy shelves around the time that you are reading this article. The manufacturer, Clinical Data, Inc, is marketing it as “a new molecular entity” and “the first and only selective serotonin reuptake inhibitor and 5-HT1A receptor partial agonist.”
Is the drug effective? The company conducted two eight-week, randomized, double-blind, placebo-controlled studies of vilazodone at a dose of 40 mg/day. One of these studies was published (Rickels K et al, J Clin Psychiatry 2009;70(3):326–333), while the other was posted on http://clinicaltrials.gov, along with less detailed results. Both studies were standard clinical trials, with the usual exclusion criteria—ie, patients were required to have at least moderate depression but no suicidal ideation, and they were excluded if they had any other psychiatric conditions, with the exceptions of generalized anxiety disorder, social phobia, and simple phobia. As usual, the devil is in the detail of the methodology of these studies, and these strict exclusion criteria limit how generalized the results are to your patients, depending on the nature of your practice.
As you can see from the table, “Vilazodone vs Placebo for Major Depression,” vilazodone yielded significantly more improvement in both depression symptom scores and response rates than placebo. These numbers are similar to those posted by other FDA-approved antidepressants.
Vilazodone appears to be an effective antidepressant, though no more effective than any of its competitors, most of which are available generically at a fraction of the cost. So why would you choose to prescribe this new antidepressant? There are two potential reasons: 1. The rumor that vilazodone has few sexual side effects; and 2. The idea that it has a unique mechanism of action.
The low sexual side effect rumor is based on the Rickels et al study. In that study, all enrolled patients were given the five item ASEX sexual dysfunction scale, and after eight weeks, there was no difference between placebo and vilazodone in changes in the score. Unfortunately, these data mean little, because most patients who entered the vilazodone trials already had sexual dysfunction (perhaps caused by the depression itself or other factors). Therefore, the only conclusion to be drawn is that neither vilazodone nor placebo made preexisting sexual problems any worse.
To understand how uninformative this study was regarding sexual side effects, let’s take a peak at a similarly designed study of two antidepressants that are well known to cause sexual dysfunction: paroxetine (Paxil) and escitalopram (Lexapro). In an eight week randomized controlled trial comparing these two drugs, the average patient began the study with a high ASEX score of 20 (the higher the score, the worse the sex). After eight weeks, the ASEX scores increased only slightly over baseline, and after 27 weeks, they actually decreased. This study did not prove that these drugs are sexually inert, it simply showed that there is a ceiling effect—at a certain level of decreased libido, nothing is going to make it any worse (Baldwin DS et al, Int Clin Psychopharmacol 2006;21(3):159-169).
Of course, the obvious way to find out if a drug causes a side effect is to prescribe it to patients who don't already have the side effect in question. This is the method used in the various studies showing that bupropion (Wellbutrin) is sexually cleaner than SSRIs. These studies enrolled only depressed patients without sexual dysfunction, randomized them to buproprion vs sertraline (Zoloft) or escitalopram, and reassessed them several weeks later. Using this methodology, bupropion vastly “outsexed” both sertraline (Segraves R et al, J Clin Psychopharm 2000;20(2):122–128) and escitalopram (Clayton AH et al, J Clin Psychiatry 2007;67(5):736–746).
The FDA agrees with our skepticism. In an email to The Carlat Psychiatry Report, Dr. Thomas Laughren, who is the director of psychiatric products at the FDA, described for us the standards the FDA requires for antidepressant makers to claim their products do not cause sexual dysfunction—and he assured us that “vilazodone has not met that standard.” Clinical Data has been officially barred by the FDA from touting vilazodone as a low sexual side effect antidepressant.
The other potential selling point of vilazodone is that it is the “one and only” antidepressant with the following dual mechanism of action: it is both an SSRI (ie, it is an antagonist) and a “5-HT1A partial agonist” which is neurotransmitter-speak for buspirone (BuSpar), the only commonly used 5-HT1A partial agonist on the market. The implication is that vilazodone is sort of like combining an SSRI and buspirone in one pill, with the putative benefits of an augmented antidepressant response and an anti-anxiety effect. This is all theoretical, of course, since there are no data showing that vilazodone is better than any other antidepressant for either anxiety or depression. We eagerly (but, suspect, futilely) await results of such future head-to-head trials! See the table: Viibryd—At a Glance
TCPR’s Verdict: Vilazodone: Effective, but be wary of low side effect claims.
Oleptro (trazodone extended-release)
Oleptro is an extended release formulation of trazodone (Desyrel). While we have come to think of trazodone as a non-addictive sleeping pill at doses ranging from 25 mg to 150 mg at night, the pill’s only actual indication is for the treatment of major depression, which it received approval for decades ago. Of course, we rarely use it for depression, because the doses required (300 mg to 400 mg, in divided doses, according to the prescribing information) would cause our patients too much sedation to make it worth using.
Which brings us to the question of why Labopharm, Oleptro’s maker, would think that a branded, expensive, extended release version of trazodone would find a place in our quiver of drugs? In other words, why would one want to experience the side effects of trazodone all day, instead of confining them to an overnight period in which one is hopefully asleep? Labopharm USA’s chief medical officer Jeff Dayno, MD, explained to TCPR that by spreading release over a 24-hour period, they hoped to decrease the sedation and allow for delivery of antidepressant dosages.
So what do the Oleptro clinical trials show? Thus far, only one randomized trial has been published (Sheehan DV et al, Psychiatry (Edgmont) 2009;6(5):20–33), and no others are listed as being in progress at http://clinicaltrials.gov. In this trial, 406 depressed outpatients were randomized to active drug or placebo. Exclusion criteria were similar to the vilazodone studies described previously, so generalizability to standard clinic patients is problematic. Titration began at 150 mg nightly, increasing every three to four days by 75 mg, up to a maximum of 375mg.
How did patients in this study fare? Oleptro outperformed placebo: starting from a baseline HAM-D-17 score of 23, Oleptro patients’ scores decreased by 11 points, versus a nine point drop for placebo. How much of a difference is two points? Big enough to be statistically significant ( P=0.012)—but is this clinically significant?
The effect size in the Oleptro study was 0.28 (not presented in the report but confirmed with Labopharm by email). A two point difference in the HAM-D scale is considered by some researchers to be clinically meaningful (for example, a 1.7 to 1.9 point change in HAM-D was defined as clinically meaningful in one review—Khan A et al, Neuropsychopharmacol 2003;28(3):552–557). But other researchers, as well as Britain’s National Institute for Clinical Effectiveness (NICE), have adopted a three-point difference on the HAM-D as the threshold for clinical significance. Therefore, depending on your definition of “significant,” Oleptro may or may not be a useful antidepressant.
What does “effect size” mean? The “effect size” is simply the difference between drug response (in this case, reduction in HAM-D scores from baseline) and placebo response. Dividing the result by the standard deviation converts the answer to a common, simple unit: effect size. Generally, in clinical research 0.2 is small, 0.5 is moderate, and >0.8 is large.
How was the drug tolerated? Daytime somnolence was the main problem, affecting 31% of the Oleptro group versus 16% of the placebo group. Other significant side effects were dizziness and nausea, occurring at roughly twice the rate of placebo. While sexual dysfunction was not thoroughly assessed, long experience with trazodone suggests that this will not be a major problem.
Interestingly, a recent small trial endorsed trazodone as potential treatment for sexual dysfunction associated with SSRIs (Stryjer R, Clin Neuropharmacol 2009;32(2):82–84). But this was only a small open trial (15 patients); the power of suggestion may have been the active ingredient, not the 5-HT2 antagonism about which the authors speculate.
So—is Oleptro worth using as an antidepressant? In patients who cannot tolerate the sexual dysfunction from most other antidepressants, and who don’t become overly sedated, Oleptro may have a place—but at about $150 for a month’s supply, insurance companies will be reluctant to foot the bill. See the table: Oleptro—At a Glance
TCPR’s Verdict: Oleptro: For most patients, not enough efficacy to justify side effects and cost, but necessary head-to-head studies with existing antidepressants are lacking.
Silenor is a branded version of doxepin, a tricyclic antidepressant that has been available in the U.S. since the 1960s or so. Doxepin is the generic name, and it is also sold by Pfizer under the brand name of Sinequan. Doxepin is such an old drug that one of its current official FDA indications is: “Psychoneurotic patients with depression and/or anxiety.”
Like other tricyclics, doxepin blocks the reuptake of norepinephrine, but it is also a very potent antihistamine, meaning that one of its main side effects is sedation. Accordingly, over the years doxepin has become a favorite non-addictive insomnia medication, usually used in very low doses (it comes as low as 10 mg).
The drug company Somaxon patented a technique for packaging generic doxepin into even smaller dosages (1 mg, 3 mg, and 6 mg), and did clinical trials showing that these work better than placebo for sleep maintenance (for example, see Scharf M et al, J Clin Psychiatry 2008;69(10):1557–1564). On the basis of these trials, they now have an FDA indication for insomnia at the 3 mg and 6 mg doses.
There’s not too much more that’s worth saying about Silenor. It would be hard to imagine a patient for whom 3 mg or 6 mg of Silenor will pose an advantage over 10 mg of generic doxepin. True, doxepin can cause a range of anticholinergic side effects, but recall that the standard doxepin dose for depression is from 75 mg/day to 150 mg/day. Given that scale, a 5 mg difference would seem unlikely to yield significantly more or less side effects.
Of course, Somaxon could have answered this question scientifically in their clinical trials if they had randomized some of their patients to generic doxepin 10 mg. Presumably they chose not to do so because they were afraid of what they would find.
A month’s supply of Silenor at 3 mg or 6 mg costs $214, according to Boston area pharmacies we contacted. On the other hand, a month’s supply of doxepin 10 mg costs $4 at Walmart. The rare patients who cannot tolerate 10 mg of doxepin can reduce the dose by opening up the capsules and mixing some of the powder in juice. The generic also comes in a 10 mg/ml liquid dose. See the table: Silenor—At a Glance
TCPR’s Verdict: Silenor: Just as good as the generic, and 50 times the price!
A “new” monoamine oxidase inhibitor (MAOI)? Well, not really. Isocarboxazid has been around since 1959. It disappeared in 1994 when the FDA called for updated data to maintain an indication. Its manufacturer, Hoffman Laroche, chose not to continue it, according to an interesting account in a voicemail response to TCPR from James Hunter, president of its new marketer, Validus Pharmaceutical. Apparently Validus spotted a commercial opportunity in reissuing Marplan as a branded drug. (According to several pharmacies contacted by TCPR, isocarboxazid is not available as a generic at this point.) The price for a 10 mg tablet averages more than $3 per pill, and given a target dose of 40 mg/day, the monthly cost could be well over $300.
Is isocarboxazid more effective than phenelzine (Nardil) or tranylcypromine (Parnate), the available generic MAOIs? Apparently not. The last comprehensive review was published 15 years ago and concluded that isocarboxazid is as effective as its competitors for outpatients, but may be ineffective for inpatients (Thase et al, Neuropsychopharmacol1995;12(3):185–219). Unfortunately, because of the lack of availability of this medication for the last decade, no more recent comparative data are available.
Are you one of the 2% of psychiatrists who actually use MAOIs regularly (Balon et al, Psychiatr Serv 1999;50(7):945–947)? If not, whichever MAOI you choose, you’ll need a handy list of foods and medications to avoid. You can download one here, or read it in TCPR, November 2006. See the table: Marplan—At a Glance
TCPR’s Verdict: Marplan: A welcome reissue, but expensive. 2006;21(3):159–169).