The pharmaceutical industry, like any large and complicated mechanism, holds itself together with an array of more or less balanced forces. They shift and change over time, finding new equilibria, but there are always factors pushing from both sides. It may look at times like something is coming in like a huge wave, sweeping away everything before it — but remember, waves roll back.

Take, for example, the early 1990s race between combichem and high-throughput screening. Depending on which hallway you were hanging around in which company (and during which month!), you might have heard the screening group expressing pride at their new-found robotic powers. In other labs, you would have heard some misgivings: was the compound collection ready to be riffled like a deck of cards? If the screening people now had these magic 96-well — 384-well! 1,536-well! — powers, shouldn’t they use them on a proportionally larger mass of compounds? You know, to increase the chances of finding something?

Then the combichem folks came rushing in to restore the balance. “Stand back!” was their advice, while the wonders of split-and-mix technology produced more compounds than anyone had ever seen in one room before. In the time it took the high-throughput screening people to format their assay, thousands — tens of thousands! hundreds of thousands! — of new compounds would have poured into the screening files. The robot arms and the resins would go around the clock, filling the shelves of the brave new repository. Order had been restored; the robot screening systems had met a worthy opponent. And indeed an equilibrium had been reached, but it wasn’t a very stable one in the end, on either side. As it turned out, there were a lot of assays that just flat out wouldn’t run very well in 1,536-well plates, at least not without major reconstructive surgery. But that was OK, because it also transpired that much of the combi-flood was junk of one sort or another — some of them not very pure and thus often not very shelf-stable, or somehow just not so druglike. The lackluster hit rates from the first combichem era ended up canceling out — well, more than canceling out, in some instances — the gains in compound numbers. Another balance of forces!

Go back a few years before combichem and you can see much the same story played out with molecular modeling. Move up a few years in the other direction, and you’ll see it with genomics, which looked for a while as if it would Utterly Change Everything. Waves of fear spread through various executive office buildings (waves of greed, too, whenever anyone took time out from the fear). Huge sums of money were thrown around by desperate companies, worried that they were missing a once-in-a-lifetime gold rush. Balance was restored in correspondingly brutal fashion, as the stocks of the genomics companies cratered with the same vigor with which they’d taken to the skies. Now, after the restoration of (comparative) sanity, it seems clear that human genomic information is simultaneously valuable and hard to interpret, a state of affairs that looks to hold up for some time.

For a more recent example, how about RNA interference, in all its guises? It’s anyone’s guess where we are in the cycle of enthusiasm, but it’s a safe bet that we’re somewhere along one. But like antisense before it, it’s likely that RNAi is going to turn out to be something less than the immediate world-changer that it might have appeared to be a few years ago. Even if it does change the world, it’s going to do it in less sudden and straightforward ways. When you consider how little we actually know about what’s going on with the various small RNAs, you have to figure that there are going to be a lot more reverses and sudden leaps before things settle down. And if you want another current example, how about stem cells? State governments are still throwing money at the area, since it’s still politically popular. But how’s all that going to look, in retrospect, if it turns out that these cells are a lot trickier and more complicated than anyone imagined? I’m afraid that we’re a long way from any kind of equilibrium in that field, and that the path to it will not be pretty.

In the chemistry end of the business, perhaps fragment-based drug discovery is a candidate. The last couple of years have seen more and more conferences, papers, review articles, and advertisements for it. But there hasn’t been a big run of fragment-based startups with bubbling stock prices, as opposed to some of the previous examples. Could that be because it’s a harder technology for people outside the field to understand, or is it just because the stock market isn’t currently in the mood for that sort of thing? At any rate, I’m relieved that fragment drug discovery isn’t being sold as yet another transformational ship’s-leaving-without-you technology, but there’s still time for that, I suppose. If someone tries to sell you on the idea, I recommend you don’t believe them: the fragment approach is going to end up right where those others have, as another wrench in the toolbox. It’ll be valuable in some cases, next to useless in others.

There’s another possible example of an equilibration in progress, this one up at the regulatory affairs level. Recently, the FDA has been notably hard-nosed on safety issues, with all sorts of delays, “approvable” letters, and outright rejections. Is this the way things are going to be from now on? Fear and despair are always on tap, so it’s easy to believe. Or is this just the first phase of the process, the over-reaction in one direction, to be followed by something from the other side? Before deciding that the industry is going to live under a permanent regulatory cloud, consider the possibility that we’re going to eventually learn some new and useful things about toxicology — or the possibility that the FDA — and who knows, even the public — might learn some new and useful things about risk assessments. I think we’re seeing the first act — maybe even the first scene — of a long and complicated production.

The point of all these cases is that there is indeed a final state of affairs that each new idea reaches — an accommodation with the various forces that made it useful or important in the first place. And this final state does not, unfortunately, correspond very well with how things might have looked during the peak of the frenzy. It’s hard to think of a technology that comes into a complex field like ours that hasn’t eventually found its place, but it’s equally hard to think of one that roared through and changed everything immediately and permanently. There’s a lot of money to be made convincing people that that’s what’s happening, of course, at any given time. It’s a human tendency to imagine that one is living at some sort of unique historic moment. But there aren’t as many of those as you might think, and they only become clear in the long view. For the shorter term, hold onto your heart (and your wallet), and you’ll likely be happier when all the dust settles.

Derek B. Lowe has been employed since 1989 in pharmaceutical drug discovery in several therapeutic areas. His blog, In the Pipeline, is an awfully good read.