Furious Seasons

September 17, 2009

Bigger Than Prozac?

There's a smallish buzz around a new drug called Valdoxan (agomelatine), an anti-depressant recently approved in Europe that's being touted as more effective than Prozac and other commonly-used anti-depressants and as having virtually no side effects. To whit, from the Mirror:

"New research shows that agomelatine - the first antidepressant in over a decade - is more effective than Prozac in treating depression. And it is not associated with some of the common side-effects of antidepressant drugs such as weight gain, sleep difficulties and sexual problems.
"A study found the £30-a-month drug, also known as Valdoxan, helped 77.7 per cent of people with severe depression compared with 68.8 per cent on Prozac. The data was presented at the European Congress of Neuropsychopharmacology in Istanbul."

That study remains unpublished and I'm skeptical of it for one reason: I've never, ever seen Prozac relieve depression in over two-thirds of patients in an eight-week trial, so it makes me wonder what was going on here. Also, there is apparently no placebo arm to the study.

Even more, a few years ago the French company Servier submitted Valdoxan for approval to European regulators and the drug was rejected as not having sufficient efficacy. The company resubmitted the drug and it was approved for use in Europe this past February (it's been submitted for approval in the UK). I've heard little about the drug since its approval.

One sign that Valdoxan is not the greatest thing since sliced bread is that Servier licensed it in the US to Novartis and the company reportedly has no plans to submit it for FDA approval until at least 2012. If it were a hotter compound, Novartis would be banging down the FDA's doors for approval.

All of that said, what is interesting is that Valdoxan is a melatonergic antidepressant as it's an agonist to melatonin receptors in the brain and it apparently works to reset human sleep patterns and circadian rhythms.

It'll be interesting to see what plays out with this drug over the next few years.

Posted by Philip Dawdy at September 17, 2009 10:12 AM

del.icio.us Digg it reddit

Comments

First: this Servier study CL3-045 was registered after completion. See

http://controlled-trials.com/ISRCTN19313268/agomelatine

So they can not even publish it, should the journals live up to their promise. And, furthermore, there could be unregistered, unpublished studies out there showing the opposite of these findings, who knows?

Oh wait, I know - there was a real, placebo- and fluoxetine-controlled efficacy trial in the EU assessment report, named CL3-022 if I remember it correctly. And, yes, it was negative. Fluoxetine aka Prozac outperformed placebo while agomelatine did not. Unregistered, and unpublished, of course.

Second: You can get the abstract of this newer trial from the ECNP 2009 Istanbul website, and there is also a paper with data from the London New Drugs Group. What does it say - they had about 500 patients in two groups, so even the small difference of less than 1,5 HAMD points was statistically significant. Clinically laughable, if You imagine the NNT!

And how did they get it? They allowed dose adjustment in the agomelatine group after 2 weeks, in the fluoxetine group only after 4 weeks. Sorry, this looks more like a joke than a scientific thing.

Third, and now it becomes interesting even to me: The marketing was to these days concentrated on the "absent side effects", which are not so absent, if we look at the newer numbers from all the posters there. It's the same rate of side effects as with SSRIs. Plus, agomelatine is liver-toxic and got approval only with security measures like regular liver function tests.

No wonder they play the "effectiveness" card now. At least in the Mirror it may work.

Posted by: PhilRS at September 17, 2009 11:36 AM

So, if I've got this right, Valdoxan is more effective than something that doesn't work?

Gee whizz: Servier sure set the efficacy bar high, didn't it? And no side effects? Just like Prozac, when it was launched. The side effects emerge subsequently, when the general population is suffering the side effects that were evident in the trials, but got hidden. As we're beginning to become aware.

Matt

Posted by: Matthew Holford at September 17, 2009 12:01 PM

Another drug to make money.
I agree with you Philip and I'm sure that with the knowledge available now about the brain and human's body it is impossible to make any difference treating any mental disease with pills.
The brain is too complex and the synaptic circuit is too complicated one of the reasons of the hideous side effects SSRIs are promoting.

Posted by: Ana at September 17, 2009 12:52 PM

Virtually no side effects? From the Valdoxan package leaflet for the user, "Common side effects: dizziness, sleepiness (somnolence), difficulty in sleeping (insomnia), migraine, headache, feeling sick (nausea), diarrhoea, constipation, upper abdominal pain, excessive sweating (hyperhidrosis), back pain, tiredness, anxiety, increased levels of liver enzymes in your blood."

Posted by: Joe at September 17, 2009 12:59 PM

their website reports on Valdoxan vs. a placebo:
http://www.valdoxan.com/index.php/in-moderate-to-severe-depression/

they really need to develop something better than HAMD for comparing treatment outcomes. Also the analysis is just wrong, using improper numeration when the data really is categorical. A better analysis would look at the frequency change in each category. But hey, we could argue about statistics forever.

Posted by: Blackeneth at September 17, 2009 06:11 PM