After many years of trying, Wyeth finally received FDA approval for its new antidepressant, Pristiq (desvenlafaxine). Don’t get too excited, though – Pristiq is simply the active metabolite of Effexor and it’s unclear whether it adds anything of value to our current quiver of antidepressants.

Initial Efficacy Studies

Thus far, three placebo-controlled trials of Pristiq have been published. The first study was a fixed dose trial in which patients were randomized to placebo or to Pristiq 100 mg, 200 mg or 400 mg/day (DeMartinis NA et al., J Clin Psychiatry 2007;68:677-688). After eight weeks, only the 100 mg and 400 mg doses were significantly superior to placebo on the Hamilton depression scale (the primary outcome). Both of these doses led to about a 46% improvement in depression symptoms after eight weeks, while placebo led to a 33% improvement over that same time period. In terms of remission rates, only the 400 mg dose bested placebo (32% vs. 19%, p=.046). The superiority of the 100 mg dose over the 200 mg dose, though, makes one question a clear dose-response curve.

The second published trial allowed researchers to titrate the dose of Pristiq up to only 200 mg/day (Liebowitz MR et al., J Clin Psychiatry 2007;68:1663-1672). At the 8 week endpoint, there was no significant difference between Pristiq (average dose, 179 to 195 mg/day) vs. placebo on either the HAM-D or the CGI-I. However, Pristiq was superior on one of the secondary measures of depression, the MADRS, and in the Visual Analog Scale for Pain. Evidently, Wyeth decided to use the VAS scale in order to challenge Cymbalta’s supremacy as the go-to antidepressant for patients with physical pain accompanying depression.

Finally, a European trial found that Pristiq 200 mg and 400 mg were both superior to placebo; I haven’t been able to obtain the full paper so I can’t report specific efficacy numbers (Septien-Velez L et al., Int Clin Psychopharmacol 2007;22(6):338-247.

Overall, these efficacy numbers are not terribly impressive, and when you look at Pristiq’s side effects at these doses, the news gets worse. In the fixed dose study, the lowest dose of Pristiq (100 mg/day) led to a 35% nausea rate, vs. 8% for placebo. Furthermore, all doses of Pristiq led to increases in both systolic and diastolic blood pressure (between 2-3% increases, depending on the dose).

Back to the Drawing Board

Because of Pristiq’s harsh side effect profile, Wyeth went back and conducted two more studies of Pristiq at 50 mg/day, in the hopes that this dose would produce fewer side effects but would still be effective. While none of this data have been published, Wyeth presented the results in summarized form in a poster at the 2007 APA meeting and in press releases. In these two studies, Pristiq 50 mg/day squeaked by placebo on the Hamilton depression scale. In the study conducted in the U.S., Pristiq decreased the HamD by 2 points more than placebo (-11.5 vs. -9.5 for placebo), and in the European study, the benefit over placebo was 2.5 points. In both of these studies, there was also a 100 mg/day arm; in the U.S., 100 mg of Pristiq was not superior to placebo, while in Europe, it was.

As Wyeth hoped, Pristiq at 50 mg was relatively well-tolerated, with nausea not being a prominent side effect in the U.S. study (though it occurred more frequently in Europe). According to Pristiq’s package insert, the 50 mg dose caused elevated supine diastolic blood pressure (SDBP) in 1.3% of patients (vs. 0.5% in placebo). In comparison, Effexor at 100 mg or less caused elevated SDBP in 1.7% of patients (vs. 2.2% in placebo) (Thase ME, J Clin Psychiatry 1998;59:502-508).

Does Pristiq Have Any Advantages over Effexor?

From a business perspective, Pristiq is essentially a patent extender for Wyeth. Effexor XR, which had blockbuster sales of $3.8 billion in 2007, is gradually losing patent protection, and Wyeth is hoping to convince psychiatrists to switch patients from Effexor to Pristiq. Should you do it?

Well, here are Wyeth’s two main arguments for Pristiq’s advantages over Effexor.

Switch-to-Pristiq Argument #1: Pristiq is easier to dose than Effexor XR.

The main Wyeth marketing point for Pristiq was spelled out in their press releases: “According to Philip Ninan, M.D., Vice President of Wyeth Medical Affairs, Neuroscience, ‘PRISTIQ is approved at a once-daily 50-mg dose that does not require titration, allowing physicians to start their patients at the recommended therapeutic dose.’” The implication is that by contrast, prescribing Effexor XR is a hassle, since you have to start at an ineffective dose of 37.5 mg or 75 mg/day and gradually increase until you get a response.

Actually, though, if you look at the original studies of Effexor you’ll find that the 75 mg dose was not ineffective, and separated from placebo fairly well. For example, in one study Effexor 75 mg led to a 3 point improvement over placebo on the HAM-D-21, comparable to Pristiq 50 mg’s 2-2.5 point improvement on the HAM-D-17 (Rudolph RL et al., J Clin Psychiatry 1998;59:116-122). A Medline search revealed three other fixed-dose studies, and all of them reported that Effexor 75 mg was significantly more effective than placebo (Khan A et al., J Clin Psychopharmacol 1998;18(1):19-25; Khan A et al., Psychopharmacol Bull;1991:27(2):141-144; Schweizer E et al., J Clin Psychopharmacol 1991;11:233-236).

There’s no question that doses of Effexor higher than 75 mg work better – but this is hardly an argument for switching to Pristiq! The bottom line is that the starting doses of both Effexor XR and Pristiq are more effective than placebo. If anything, Effexor poses a significant advantage, because you can reliably get a better response when you increase the dose. The data on Pristiq, on the other hand, show no clear dose-response relationship at all.

Switch-to-Pristiq Argument #2: Pristiq has fewer drug-drug interactions.

Effexor is metabolized primarily by the P450 2D6 liver enzyme, with some contribution of 3A4. According to the package insert, however, combining Effexor with drugs that inhibit these enzymes is unlikely to be clinically significant, because Effexor is metabolized to an equally pharmacologically active compound, namely, desvenlaxine, or Pristiq. Because of this, when you inhibit Effexor’s metabolism, the blood levels of Effexor rise but blood levels of Pristiq decrease, with the net result being that nothing significant, either positive or negative, is likely to occur. In addition to being metabolized by 2D6, Effexor is considered to be a “weak” inhibitor of 2D6, and is unlikely to cause other drug levels to rise significantly (see Sandson NB, Drug Interactions Casebook, American Psychiatric Press, 2003). While Pristiq is not metabolized via the P450 enzymes, it is still metabolized in the liver, mainly via glucuronidation, and drug interactions are unlikely.

What about patients with liver or kidney disease? You have to decrease the dose of both drugs in these cases, so there’s no advantage for Pristiq there. In fact, the FDA asked Wyeth to submit additional safety information because of unspecified “serious adverse cardiovascular and hepatic effects” that occurred during trials for easing hot flashes in menopausal women (http://www.reuters.com/article/ health-SP/idUSN2442193420070725). Presumably, these side effects occur only at doses higher than 50 mg. Of note, the Pristiq package insert states that “dose escalation above 100 mg/day is not recommended” in patients with hepatic impairment, which sounds a little ominous. I assume we’ll find out more details over time.

So, what to make of poor Pristiq? As compared to Effexor, it has no greater efficacy, it is no easier to dose (in fact, a little harder, because of the lack of a dose-response relationship), it is no better tolerated at comparable doses, and it has no meaningful drug interaction advantage.

TCPR Verdict:

Pristiq: Not a novel antidepressant.