There is a new blog entry by Jonah Lehrer, “How Does Prozac Work?” (Nov 17, find it here) that touches on the decline effect in pharmaceuticals. It’s not good news. And in a strange way it is stressing me out personally.
First Lehrer discusses the failure of a promising new anti-depressant drug, designated GSK372475, from Glaxo. The drug seemed to have potential to boost serotonin and dopamine levels substantially. Researchers see a link between serotonin, dopamine, and depression. But in two clinical trials, the placebo patients were actually better off — not only less depressed, but less subject to nasty side effects like nausea and insomnia. Oh well, millions of dollars for nothing.
But then Lehrer drops the real bad news: After years of frustration and failed trials of this type, the pharmaceutical companies are cutting their losses and simply spending less on basic research into the causes of depression. Lehrer writes:
“Given such struggles, it’s not surprising that drug companies are dramatically scaling back research into the brain. (Most recently, four leading drug firms, including Merck and GlaxoSmithKline, announced that they were cutting neuroscience R&D. They cited the disconnect between scientific funding and pharmaceutical success.)”
Now why is this stressing me out? What is the connection?
Lehrer reported last year on the increasingly common phenomenon of drugs doing well in initial small-scale trials, then failing when the trials were expanded. There are people in the pharmaceutical industry, senior statisticians, who are quietly worrying that there is some kind of failure in the scientific method. The small-scale trials are sufficiently large to establish a positive drug effect, yet when more people take the drug, we get a weaker effect. No one knows why. (Well, I think I do. But apart from me, not many people even worry about this issue. As I’ve said before, most recently last Sunday, Jonathan Schooler is one of the few academics brave or foolhardy enough to consider the possibility that the universe just likes to decline.)
What worries me at the moment is that the statisticians who work for Big Pharma are giving up too easily. Glaxo management specifically linked the cut in funding to the failure of large Phase III trials. They can’t figure out why something looks good in the lab, when applied to the first few subjects, and then gradually stops looking good.
These failures could be explained, or anyway far better understood and forecast, using maximum entropy methods and the framework laid down by Edwin Jaynes. But first someone (me) has to write a book about how to do this, and get it published, and persuade people to read it.
Meanwhile, the consequences of failure are backing up into basic R & D. The neuroscientists come up with theories, but then their theories fail to translate into successful drug trials. So Big Pharma cuts funding to neuroscience.
At the moment, I feel quite irrationally responsible for those neuroscientists losing their grants. I need to get my damn book out there, so Big Pharma will have a better explanation for its failed drug trials, and not take out its frustrations on brain research.