Cymbalta for GAD: Repeating the Pattern

Lilly's drug Cymbalta (duloxetine, aka Yentreve) was approved by the FDA as a treatment for generalized anxiety disorder today.

Is this good for people with GAD? Is this good for Lilly? Are there any academic cheerleaders in the house?

People with GAD: Sometimes a medication really does a good job in treating symptoms and sometimes there are red flags that the science does not match the marketing. In this case, it took very little digging to find a red flag – just read the press release!

In clinical trials, on average, patients treated with Cymbalta for generalized anxiety disorder experienced a 46 percent improvement in anxiety symptoms compared to 32 percent for those who took placebo, as measured by the Hamilton Anxiety Scale [HAM-A].

I have not seen the full text of what appears to be the only published study of Cymbalta for GAD. I cannot find the other two duloxetine studies on Medline (the press release claims there have been three studies), which seems odd. Perhaps I should dig a little deeper… In any case, let’s move on to the efficacy of duloxetine in the study referenced in the press release.

My guess (from a similar study examining Zoloft for GAD) is that participants averaged about 25 on the HAM-A at the start of treatment. Using the press release’s numbers, the Cymbalta group comes down to an average of 13.5 while the placebo group came down to an average of 17. Is a difference of 3.5 points on the HAM-A noticeable? Could a clinician tell the difference between someone with a 17 and a 13.5 on this measure? I think that in most cases, yes. But it is probably a small difference.

Adverse events reporting in press releases (and often in journal reports) is a joke, so I am not sure about the safety of the compound. The infamous Traci Johnson case suggests that there may be a risk of suicidality with Cymbalta, but I would prefer to see more data before making a firm association. Of course, given the general link between SSRIs (and newer antidepressants) and increased suicidality, my initial instinct is to be leery of Cymbalta in this regard.

Good for Lilly: This is great news for Lilly. As the Zyprexa debacle continues to unfold, Lilly’s search for good news appears to have arrived in Cymbalta. According to the Indianapolis Star, Cymbalta sales hit $1.3 billion in 2006. With the GAD indication and its rapid expansion into the depression market, I’m guessing that we’re looking at least a $2 billion drug for 2007. Not bad for a drug that has no evidence of working better than any of the older, now mostly off-patent SSRIs.

Academics in the House: Dr. Susan Kornstein of Virginia Commonwealth University, whom I have mentioned in the past in relation to her highly positive statements about Lexapro, is back. In the press release, she states:

With this approval, physicians and patients will be happy to know that there is another medication now available to treat this debilitating condition.

Yeah, they’ll be dancing in the streets. With all SSRIs likely showing similar efficacy to Cymbalta and with psychotherapy showing pretty good (but not great) outcomes for GAD (likely better in the long-term than with meds), we needed a more expensive medication because… WHY?

It is amazing how the pattern repeats itself. Show a medication is at least a bit better than a placebo in the short-term, have an “independent” academic give it the stamp of approval, then talk about how it is an important new addition that will prevent untold amounts of grief and suffering. The safety data are often glossed over and the benefits compared to existing treatments are not discussed at all. This helps people (outside of the sponsoring company)… HOW?