Column: In the pipeline

The recent row over antidepressants reminds us how little we know about the brain, says Derek Lowe

You’ve probably read the recent headlines about a study which seems to show that many antidepressant drugs barely outperform placebos (see p19). If you missed it, though, don’t worry - there’ll be another coming along in another year or two. The story has done the rounds before on roughly that schedule, and shows no signs of going away. 

That’s because there’s a lot of arguing room inside the subject. Critics of the latest work point out that it only considered shorter clinical studies that were less likely to show robust effects. Central nervous system (CNS) drugs are notorious for slow onsets of action, sometimes taking months to have an effect. This makes studying them in a clinical setting quite difficult, and quite expensive. 

But, on the other hand, CNS drug trials are also notorious for showing very high placebo response rates. Antidepressants probably have the highest of all, since the hope provided by participating in a clinical trial of a new drug is, for many patients, a valuable therapy in itself. Up to 40 per cent of the patients getting cryptically labelled corn starch can show improvement in their depressive symptoms. That doesn’t make running the drug trials any easier, either. 

Chemical hope 

It’s understandable to feel disappointment at the thought that these expensive medicines can actually face stiff competition from sugar pills. But that mindset assumes that the placebos are doing nothing, which is a mistake: the improvement in depression is absolutely real.  

If depression is indeed a chemical process in the brain (and it’s hard to imagine what else it could be) then we have to be ready to deal with the many ways that brain chemistry can be altered. No one doubts that fear sets off profound reactions and chemical changes throughout the body and the brain, changes that can be triggered by all sorts of seemingly minor stimuli. Why shouldn’t hope work the same way? 

The problem is that hope, like fear, can wear off in time, which might be why a shorter study could suffer more from the placebo effect than a longer one - and why in the long run, these drugs could really be helping some patients. It could even be that the drug’s effects take over gradually as the placebo effect diminishes, although I’d hate to try designing a trial to shore up that hypothesis. None of this is the stuff of catchy headlines, though: ’Statistics Suggest That Antidepressants Help Eventually, But Maybe Not At First’ is unlikely to appear in your daily newspaper anytime soon. 

Bewildering cascades  

But there are complications on the biochemistry end of the argument, too. The biggest one is that brain chemistry is so ridiculously complex that we frankly have only a vague idea of how any of these drugs work at all. Our best CNS compounds are very crude tools indeed and, with any luck, future generations will roll their eyes in disbelief at them.  

Even the more selective compounds (such as the serotonin reuptake inhibitors) set off bewildering physiological cascades. Some of the other CNS drugs, especially those for schizophrenia, hit so many targets at once that it’s impossible to say exactly what’s going on. And then there’s patient-to-patient variation, which any psychiatrist will tell you is never to be ignored. 

So what lessons should we draw from this latest flap? Well, it’s worth remembering that there’s a bias in the popular press for ’Everything You Know Is Wrong’ stories. There’s also a bias for narrative structures, particularly ones involving good guys and bad guys.  

If we don’t want to always end up in the latter category, we in the drug business need to show some proper humility when measuring ourselves against the complexity of the human body and the human brain. It’s understandable that we should tout the efficacy of our drugs - but a dose of realism might do us some good. 

But would it do depressed patients any good? Here’s a disturbing thought for you - if the placebo effect is a real component of taking these drugs, will it still come through if people start to think that the drugs are only as effective as a placebo? 

Derek Lowe is a medicinal chemist working on preclinical drug discovery in the US