When the drugs don't work

I've just taken my medication, with a glass of wine. I'm reminded of a comment I heard by a very senior pharmaceutical chief at a small meeting I was involved with in my previous life - held in Basle, Switzerland, with luminaries such as Peter Goodfellow and Craig Venter in attendance. It was looking at genomics. Allen Roses from GSK suggested that the vast majority of drugs only work in perhaps 30-50% of people (see hBBC report here).

Why? It's because of individual differences. We're all wired slightly different. While Roses may have been exaggerating a little (it would be fairer to say that there are a number of people in whom drugs have no effect, but that the effect of most drugs varies quite a bit among others, making it hard to get doses right), he does have a point.

I'm fascinated by individual differences, especially where they relate to the perception of wine. I've mentioned before that a winewriter I greatly respect loves Springfield's Life From Stone Sauvignon Blanc; I hate it - I think the wall of pure, unripe, methoxypyrazine character this wine presents to me is verging on a fault.

But Roses' point is that if we can get a handle on peoples' genetic make up, for example by cheap SNP arrays, we can prescribe drugs much more accurately. Personalized medicine. We can also rescue useful drugs that have failed clinical trials because of their adverse effects in small groups.