A notable year for closures and layoffs

I’ve been blogging about the drug industry for nearly five years now, and a clear pattern has emerged when I get to the end-of-year retrospectives. Unfortunately, it’s not a cheerful one. So far, every single one of them has begun with a sigh of ’At last that’s over with’, and I’m not about to break that streak this year. 

How could I? 2007 was notable for a series of site closures and layoffs across the industry. If you’ve been around the business for a while, you’ll surely drop this year in the back of the bottom file drawer; if you’re new, all I can say is that they’re not all supposed to be like this.  

And since the patent expirations are still coming right along, and since we don’t seem to have made any obvious recent breakthroughs, it’s easy to be gloomy when you think about the prospects for 2008. 

Rather too easy. Many workplaces have someone who sits in the back of the conference room and seemingly does nothing but tell people that their latest ideas aren’t going to work. Naturally, gloomy predictions have a high success rate, since most new ideas don’t in fact work. And if I wanted to, I could probably dust off the same drug-industry forecast every year: rain and snow, punctuated by earthquakes and mudslides. Most years, I wouldn’t be very far wrong.  

But always taking the safest route to being right is no way to live. Perversely enough, the rash of ugly, expensive Phase III clinical failures we’ve been experiencing in the last few years might actually be a good sign. What these say to me is that perhaps we’re not failing for the same basic and obvious reasons that we used to. No, I think we’re moving along to failing for tricky and complicated reasons.  

Over the years, our clinical failures may have been gradually shifting further down the development path. Of course, we still have preclinical problems. But once into trials, I wonder if an increasing proportion of compounds might be making it further along than they used to. In a survey a few years ago, it appeared that pharmacokinetic difficulties weren’t killing off as many compounds as they did twenty years before, for example, and that would mean that we get past Phase I more often. Efficacy failures had moved up into the first row of reasons for failure, and that’s primarily the sort of thing you find out in Phase II. It may well be that toxicity is now moving up as well, but not the sort of toxicity that sends you out of the clinic early - it’s the sort that shows up later, in large populations with longer dosing: Phase III. 

If that’s true, then we can hope that the process will continue. As I’ve pointed out here before, we don’t need much of an improvement in clinical success rates to make a big difference. This will require us to learn more about toxicity. Too often we don’t see tox problems coming and don’t understand what’s happening when they arrive. But this might be changing. Things that were black boxes a few years ago are slowly becoming understood, such as the widespread cardiac problem of QT interval prolongation. We definitely don’t have that figured out yet, but we’re a lot better off than we used to be. 

As we start to get a handle on these factors, projects are naturally going to be killed off by them. But that’s what predictive assays are supposed to do - and that would certainly be a lot more enjoyable than our present system of killing them off after years of effort. 

Perhaps this is just the awkward age of drug discovery. We know enough to get deep into the clinic, but not enough to reliably come out the other side. That makes this a particularly difficult and frustrating age (not to mention an expensive one), but if I’m right, it’s also not a permanent one. 

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