Friday, October 01, 2010

Cymbalta and Effexor: Hype Over Science

ResearchBlogging.org
Remember the hype around the serotonin-norepinephrine reuptake inhibitors (SNRIs)? Effexor and Cymbalta impact both serotonin and norepinephrine, so they should be more effective than SSRI’s in treating depression? Mind you, that’s not a high bar to clear - it’s not like SSRI’s are much better than placebo. So get the hell outta the way, Prozac and Paxil, because Cymbalta and Effexor will unleash their incredible efficacy onto the world of psychiatry. Doubt me? Read this 2009 article regarding the wonders of Pristiq (son of Effexor) and learn about how “The emergence of the selective serotonin reuptake inhibitor (SSRI) and serotonin norepinephrine reuptake inhibitors (SNRI) antidepressants has improved the treatment of MDD.” Or this press release from Wyeth. Or Dr. Danny Carlat’s experience selling Effexor to his peers. I don’t think anyone who has followed drug marketing would deny that both Wyeth and Lilly tried to pimp Effexor and Cymbalta as working better because of their SNRI properties.

But is that actually true? A team of German researchers examined the data and concluded that neither Effexor nor Cymbalta really work better than SSRIs. They actually found a small advantage for Effexor over SSRIs for treatment response (but not depression remission), but they also found that the manufacturer was hiding studies from them (and the rest of the world). I haven’t said this for a while, but enter Charles Nemeroff. To understand the research by the Germans, we first need to recall that a 2008 study (lead author: Nemeroff) found

...the pooled effect size across all comparisons of venlafaxine versus SSRIs reflected an average difference in remission rates of 5.9%, which reflected a NNT of 17 (1/.059), that is, one would expect to treat approximately 17 patients with venlafaxine to see one more success than if all had been treated with another SSRI. Although this difference was reliable and would be important if applied to populations of depressed patients, it is also true that it is modest and might not be noticed by busy clinicians in everyday practice. Nonetheless, an NNT of 17 may be of public health relevance given the large number of patients treated for depression and the significant burden of illness associated with this disorder. [my emphasis]


As I wrote then, the benefit to public health claim is ridiculous. To understand the reasons why this is so laughable, please check out my prior post on the topic. This meta-analysis included a bunch of data from Wyeth that was previously unpublished...

Which leads to the freshly published meta-analysis on how Effexor compares to SSRIs. The German researchers requested unpublished data from Wyeth and only got some of it - you’d think that just maybe Wyeth sent them the “good news” data and maybe held back on some of the “bad news” data. So when an ever-so-small benefit emerged for Effexor (5% high treatment response rate), well, call me crazy, but I ignored it. We’re not playing with a full dataset because the manufacturer wants to keep some of it hidden, so shame on Wyeth and let’s look at Effexor with a little bit of suspicion. So Effexor vs. SSRIs - no difference. Except that more people drop out of clinical trials on Effexor due to side effects compared to SSRIs (about 3% more). So even if you believe that Wyeth’s hidden data really doesn’t impact these findings, we’re left with a very small advantage for Effexor that is probably negated by its slightly higher dropout rates.

Cymbalta. It had a 3% higher dropout rate due to adverse events and the same efficacy as SSRIs. So nothing to write home about, except that it costs a boatload more than generic SSRIs and is harder to tolerate. But Cymbalta has been marketed to the gills and is clearing $3 billion a year in sales. Hey, this is the company marketed Zyprexa for dementia (oops), and for, well, lots of other stuff (1, 2). So it’s not surprising at all that they can take a mediocre antidepressant like Cymbalta and turn it into a big moneymaker - the wonders of a good marketing department. But Depression Hurts and Cymbalta is a painkiller. Well, that’s fine and dandy until you actually look at the data which show Cymbalta doesn’t do much for pain in depression.

It’s time to get over the hype surrounding SNRIs. The next “advance” in antidepressants, well, who knows what it will be - but let’s hope it’s something a little more substantial than SNRIs. But I’m not hopeful. And no, I don’t want to hear anything more about agomelatine.

I know it’s been a long time between posts. So pardon me if my writing is more awful than usual. And it doesn’t mean I will be posting regularly. Thanks to the multiple readers who sent me a copy of this article.

Citation to new meta-analysis of Effexor and Cymbalta:
Schueler, Y., Koesters, M., Wieseler, B., Grouven, U., Kromp, M., Kerekes, M., Kreis, J., Kaiser, T., Becker, T., & Weinmann, S. (2010). A systematic review of duloxetine and venlafaxine in major depression, including unpublished data Acta Psychiatrica Scandinavica DOI: 10.1111/j.1600-0447.2010.01599.x