There’s an idea that’s been floating around the drug industry for some years now. It goes in and out of focus, but it refuses to completely die: Why not ditch the research labs and just in-license all your compounds? It’s been at least 10 or 12 years since I first heard of this plan (although I’m sure it’s an older idea than that), and it still keeps popping up as a new thought experiment.

“Wait,” you might be saying, “don’t we already have a company whose sales pretty much all come from outside compounds? Big outfit, name of Pfizer?” Point taken, but remember, Pfizer has lots of research labs as well, whose contents they rearrange every couple of years when some sort of gong goes off in Groton, CT. (That’s my theory anyway; it fits the facts about as well as any other.) They do indeed bring in most of their revenues from outside compounds, but they haven’t given up on trying to discover their own. Yet.

Will they? Will anyone? This brings up my first objection to the whole idea. You’d have to figure that these numbers have been run before, and found wanting. I find it hard to believe that no one’s seriously taken a look. The idea of cutting the discovery organization loose is very appealing – well, not to me, naturally, but to the accountants. The research labs are a whirlpool of red ink, with no guarantee of anything ever coming back out, and I’m sure that the master spreadsheet perks up considerably when that section is deleted. Since no one’s quite had the nerve to do that yet, there must be some other concerns, and they must be pretty substantial ones.

One of those could be the sheer number of compounds available. Keep in mind that everyone else is looking for compounds to bring in, too. Is there enough plankton out there to keep an entire whale swimming around? I’m not so sure about that. You can run thought experiments retroactively and make this sort of thing work out – if you’d just licensed in that one over there, and then this one over here – but doing that moving forward is a different matter. The question is, is the rest of the industry capable of putting up enough potential drug candidates, compared to your own internal efforts?

This would probably vary from one therapeutic area to another, but not in a way that you might be able to take advantage of. That’s because, at least as far as I can see, there’s no particular area that keeps running a surplus of potential clinical candidates. Your virtual company would presumably have to have clinical and regulatory expertise in a few defined areas, and perhaps you could set that up based on the current landscape of potential compounds. But that landscape isn’t fixed. A few years later, you might find yourself with little to bring in and, since you’re depending on other people to do the discovery work, there’s not much you can do about that.

There are other potential money problems. Your own labs produce things that you can keep to yourself, and that you’ve already paid for. You can budget for that stuff in fairly coherent fashion. But the cost structure of an all-in-licensed company would be a bit jumpy, because the cost of doing all those deals would be, too. No one does these things in a vacuum, and any compound that’s really worth bringing in will surely have some other people looking it over, too. And what happens when Company X announces that from now on, it’s going to survive purely on outside compounds? Well, I’ve always thought that it means that the price of deals, around the whole industry, just goes right up.

After all, everyone else now knows that Company X has no choice but to make deals at some point. Why not make them pay for it, if you’re a competitor? Or make them pay you, if you’re a potential partner? This situation would become especially acute if the company ever ran a large air bubble in its drug pipeline, which everyone does at some point. Then the poker game would get ever nastier, and the old saying would come into force: “You need a deal real bad? Well, here’s a really bad deal!” At some point, you’d have to wonder if you weren’t paying as much (or more) for things than if you’d found them yourself.

So, taking all of these things together, I don’t see anyone trying this plan on any sort of large scale. The mix of internal and external compounds will keep shifting at different companies, but I think that a large company is always going to have to have that home-made component. In a way, the in-house research serves as sort of a buffer, compensating for the wild swings in pH that would occur if a company tried to go all external. And if my poker-game speculation above is accurate, there may be a signaling component, too: an internal research effort tells other companies that you always have your own resources to fall back on.

The only way I can make this come close to working might be in oncology. There are an awful lot of potential candidates out there – many of them, mind you, not quite as well thought out as they could be – but the field does have a history of crazy ideas that actually worked out. Think about the revival of thalidomide, or arsenic trioxide. Any therapeutic area that can accommodate those guys clearly has a lot of leg-room in it. And there’s the advantage that the clinical trials don’t have to be as humongous as in some other areas. Also, oncology research centers tend to bring in all sorts of patients with all sorts of cancers, centralizing your potential clinical trial cohorts (not to mention your eventual customers). No, you’d have to be off your head to try something like this in CNS or cardiovascular, but it might have a shot in oncology.

Perhaps a portfolio of wild-eyed cancer ideas, brought in from all over the place, might be able to pay off. The thinking here is similar to junk bond funds, where the pitch was that individual risks of default, while real, could be mitigated by portfolio analysis when you brought enough different types of high-yield bonds together. If you wanted to go further with the analogy, a pipeline of compounds like this might be funded by breaking things out into risk tranches, just like they do in the bond industry: people who funded the longer shots would be promised a higher share of the profits, and so on.

That would depend on being able to accurately assess risks, though, which isn’t easy in this business. But hey, as we’ve been proving for some months now, it turns out not to be as easy as it looks in the regular bond industry, either. How much worse could we be at it? Maybe the question isn’t “How come no one tries bringing in all their compounds from outside?”

It’s, “How come no one tries bringing in all their money from outside?”

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.