Tuesday, June 8, 2010

Peak Drug Discoveries?

Now, I have heard it all. First, we started with Malthus and peak food, then there was peak oil and peak gold. Now Megan McArdle declares drug discoveries have peaked:
I have an article in this month's magazine about pharmaceutical pipelines.  It sounds dry (and perhaps the piece is, a bit), but it's incredibly important.  Ever since what looks like a historical peak in the mid nineties, the number of new medical entities being approved by the FDA has been in a long, slow decline.

This is, obviously, a good place for a typical libertarian jeremiad against the Food and Drug Administration, but while I think there are quite a lot of problems with the way that that body operates, it cannot bear all of the blame for the decline.  As I discuss at some length in the feature, it may just be getting harder to find new drugs.  There are a lot of reasons for this, but all of them add up to a slower pace of drug discovery.
Since medical discoveries are just that "discoveries," we just don't know what we are not seeing until a discovery is made. There is thus no way to no the size and quality of such advances. McArdle simply doesn't understand the nature of discovery, or she is inhaling too many inside the beltway fumes to allow her mind to put two and two together.

It shouldn't come as a surprise that McArdle's discovery of fewer drug discoveries allows her to leap the possibility that Obamacare is a good thing (though she's really not sure--give her a few years, she says):

Obviously, one of the questions this raises, for me and long time readers of my blog, is whether this changes how I feel about health care reform. That's a complicated question, for a lot of reasons--for starters, watching the adminstration [sic]and the state of Massachusetts fool around with quasi-controlling insurance prices, I now think that I may have seriously underestimated the risk that this thing will turn into a massive fiscal disaster that could precipitate a serious budget crisis. (Of course, it's a work in progress, so ask me again in a few years). But obviously, one of my major objections has been that the program would slow the rate of medical innovation; if medical innovation is slowing anyway, maybe that's not such a big problem....I'm not sure what to think, actually, for a couple of reasons.
Of course, the reality is that drug discoveries will decline, not because we have reached some kind of nutty "Peak Drug Discoveries" but because Obamacare is all about disincentives. With less rewards for discovering new drugs, there will be less incentive to try and find such drugs and less incentive to invest in drug discovery research.


  1. While not a true "catch 22" for pharmaceutical companies they are in a dilemma. First- their patents expire, potentially capping the cash flows generated by a particular drug. Second- new drugs are expensive to bring to market and their ability to recoup costs is under attack. Third- Everyone is chasing the same "global" diseases (diseases that a larger proportion of the population are likely to get). And fourth- at some point the development costs exceed the revenue generated by the drug.

    Capital needs to be allocated to those products which will give them the best bang for their investment dollar. First you go for big things like allergies and other "reasonably occurring" non gender specific conditions. Then you move to those with a relatively large demand (heart disease? obesity?), then you move down the list to the smaller (more rare) conditions as appropriate and you finally end up with those where the investment is not worth the reward.

    So if your "big guns" only give you coverage for a certain amount of time, you don't have pricing power anymore to ensure you recoup your investments, you don't have the same population of "wide use" drugs and the only "big" cures left have potentially disproportionate ratios of revenue to development cost, what do you do?

  2. Why shouldn't development costs decrease as companies gain experience in finding cures and begin to target diseases affecting smaller groups? If the group is smaller the variety of potential causes is probably also smaller, and thus, it could be easier to find a cure.

  3. If you want to see innovation, get rid of the FDA, and let people be free to decide what drugs they want. At worst, make the FDA optional.

  4. I see two reasons why costs won't automatically decrease and one why they would. First, certain development costs are fixed due to regulatory constraints (e.g. the FDA) and the process for getting approval of a drug with 50,000 potential customers is the same as one for 100 million so you won't see savings there.
    Second- the experience and learning curves are probably not the same as in a manufacturing environment. Much of the cost is in R&D where you may or may not be able to leverage prior work; if you get it then it is great but if not... oh well. The research process, I suspect (I don't KNOW), is akin to a scientist going for there PhD. Lots of trial and error and study. Many graduate students I have met have had large bumps along the road to success. I imagine this process is similar.

    One are where you would see decreased expenses is in manufacturing efficiencies and scale. But that won't solve everything.