Corante

About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: derekb.lowe@gmail.com Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Emolecules
ChemSpider
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
PubChem
Not Voodoo
DailyMed
Druglib
Clinicaltrials.gov

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
Kilomentor
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
ChemBark
Realizations in Biostatistics
Chemjobber
Pharmalot
ChemSpider Blog
Pharmagossip
Med-Chemist
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
SimBioSys
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Business|Bytes|Genes|Molecules
Eye on FDA
Chemical Forums
Depth-First
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa


Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
FuturePundit
Aetiology
Gene Expression (I)
Gene Expression (II)
Sciencebase
Pharyngula
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net


Medical Blogs
DB's Medical Rants
Science-Based Medicine
GruntDoc
Respectful Insolence
Diabetes Mine


Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem


Politics / Current Events
Virginia Postrel
Instapundit
Belmont Club
Mickey Kaus


Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Unburied | Main | How Much Basic Research? »

September 9, 2004

How It Really Works

Email This Entry

Posted by Derek

So is this the attitude we're up against? Here's a thread on Slashdot on the clinical trial disclosure issue - titled, I note in light of yesterday's post, "Medical Journals Fight Burying of Inconvenient Research". My favorite verb again! The comments range from the insightful to the insipid (for another good reaction to the clinical trial controversy, go here.)

A comment to the original Slashdot item disparages the idea that NIH is the immediate source of all drugs, and recommends reading my site, both of which actions I appreciate. But the first response to that was:

"No, (NIH-funded labs) just do the basic research that results in the drug leads. The companies then do the expensive but scientifically easy trials and rake in all the money (and now it seems, the credit as well)."

Wrong as can be, and in several directions at once. In a comment below, blogger Sebastian Holsclaw urges that we take this kind of talk seriously because it's more widespread than we think. I'm afraid that he might be right. The problem is that many people don't seem to understand what it is that people like me do for a living. I think that there must be plenty who don't even grasp how science works in general. Allow me to go on for a while to explain the process - I'd appreciate any help readers can provide in herding the sceptics over to read it.

Try this: If Lab C discovers that the DooDah kinase (a name I whose actual use I expect any day now) is important in the cell cycle, and Lab D then profiles its over-expression in various cancer cell lines, you can expect that drug companies will take a look at it as a target. Now, the first thing we'll do is try to replicate some of the data to see if we believe it. I hope that I'm not going to shock anyone by noting that not all of these literature reports pan out.

But let's assume that they do this time, making DooDah a possible cancer target. What then? If we decide that the heavy lifting has been done by the NIH-funded labs C and D, then what do we have so far? We have a couple of papers in the Journal of Biological Chemistry (or, if the authors are really lucky, Cell) that, put together, say that DooDah kinase is a possible cancer target. How many terminally ill patients will be helped by this, would you say? Perhaps they can read about these interesting in vitro results on their deathbeds?

What will happen from this point? Labs C or D may go on to try to see what else the kinase interacts with and how it might be regulated. What they will not do is try to provide a drug lead, by which I mean a lead compound, a chemical starting point for something that might one day be a drug. That's not the business these labs are in. They're not equipped to do it and they don't know how.

(Note added after original post): This is where the drug industry comes in. We will try to find such a lead and see if we can turn it into a drug. If you believe that all of what follows still belongs to the NIH because they funded the original work on the kinase, then ask yourself this: who funded the work that led to the tools that Labs C and D used? What about Lab B, who refined the way to look at the tumor cell lines for kinase activity and expression? Or Lab A, the folks that discovered DooDah kinase in the first place twenty-five years ago, but didn't know what it could possibly be doing? These things end up scattered across countries and companies. And all of these built on still earlier work, as all the work that comes after what I describe will build on it in turn. That's science, and it's all connected.

Here in a drug company, we will express the kinase protein - and likely as not we'll have to figure out on our own how to produce active enzyme in a reasonably pure form - and we'll screen it against millions of our own compounds in our files. We'll develop the assay for doing that, and as you can imagine, it's usually quite different than what you'd do by hand on the benchtop. Then we'll evaluate the chemical structures that seemed to inhibit the kinase and see what we can make of them.

Sometimes nothing hits. Sometimes a host of unrelated garbage hits. For kinases, these days, these usually aren't the case - owing to medicinal chemistry breakthroughs achieved by various drug companies, let me add. So if we get some usable chemical matter, then I and my fellow med-chemists take over, modifying the initial lead to make it more potent, to increase its blood levels and plasma half-life when dosed in animal models, to optimize its clearance (metabolism by the liver, etc.), and make it selective for only the target (or targets) we want it to hit. Often there are toxic effects for reasons we don't understand, so we have to feel our way out of those with new structures, while preserving all the other good qualities. It would help a great deal if the compounds exist in a form that's suitable for making into a tablet, and if they're stable to heat, air, and light. They need to be something that can be produced by the ton, if need be. And at the same time, these all have to be structures that no one else has ever described in the history of organic chemistry. To put it very delicately, not all of these goals are necessarily compatible.

I would love to be told how any of this comes from the NIH.

Now the real work begins. If we manage to produce a compound that does everything we want, which is something we only can be sure of after trying it in every model of the disease that you trust, then we put it into two-week toxicity testing in animals. Then we test in more (and larger) animals. Then we dose them for about three months. Large whopping batchs of the compound have to be prepared for all this, and every one of them has to be exactly the same, which is no small feat. If we still haven't found toxicity problems, which is a decision based on gross observations, blood chemistry, and careful microscopic examination of every tissue we can think of, then the compound gets considered for human trials. We're a year or two past the time we've picked the compound by now, depending on how difficult the synthesis was and how tricky the animal work turned out to be. No sign of the NIH.

The regulatory filing for an Investigational New Drug needs to be seen to be appreciated. It's nothing compared to the final filing (NDA) for approval to market (we're still years and years away from that at this point), but it's substantial. The clinical trials start, cautiously, in normal volunteers at low doses, just to see if the blood levels of the compound are what we think, and to make sure that there's no crazy effect that only shows up in humans. Then we move up in dose, bit by bit, hoping that nothing really bad happens. If we make it through that, then it's time to spend some real time and money in Phase II.

Sick patients now take the drug, in small groups at first, then larger ones. Designing a study like this is not easy, because you want to be damn sure that you're going to be able to answer the question you set out to. (And you'd better be asking the right question, too!) Rounding up the patients isn't trivial, either - at the moment, for example, there are not enough breast cancer patients in the entire country to fill out all the clinical trials for the cancer drugs in development to treat it. Phase II goes on for years.

If we make it through that, then we go on to Phase III: much, much larger trials under much more real-world conditions (different kinds of patients who may be undergoing other therapy, etc.) The amount of money spent here outclasses everything that came before. You can lose a few years here and never feel them go by - the money that you're spending, though, you can feel. And then, finally, there's regulatory approval and its truckload of paperwork and months/years of further wrangling and waiting. The NIH does not assist us here, either.

None of this is the province of academic labs. None of it is easy, none of it is obvious, none of it is trivial, and not one bit of it comes cheap. We're spending our own money on the whole thing, betting that we can make it through. And if the idea doesn't work? If the drug dies in Phase II, or, God help us all, in Phase III? What do we do? We eat the expense, is what we do. That's our cost of doing business. We do not bill the NIH for our time.

And then we go do it again.

Comments (28) + TrackBacks (0) | Category: Academia (vs. Industry) | Clinical Trials | Drug Development


COMMENTS

1. Chris on September 10, 2004 1:13 AM writes...

Bravo. Very succinct and thorough explanation of the discovery and development phases of pharmaceuticals. I've tried millions of times to explain to people that the NIH only funds shots in the dark. Theres a LOT of work that needs to be done between a shot in the dark and a verifiable treatment. Apparently people think the step between in vitro simple cell studies and human treatments is but a small hop and a skip. My explanations were always long, convoluted, and required moderate scientific knowledge to understand. You're explanation is very short and simple compared to mine. Thank you Thank you Thank you Thank you.

Permalink to Comment

2. Uli on September 10, 2004 9:05 AM writes...

In a major address at Duke University on September 7, Schering-Plough CEO Fred Hassan presented a four-point plan for sweeping health policy reform for the United States that he said should be adopted by whichever candidate wins the November presidential election.


Related to Derek's entry on "How it Really Works", Hassan emphasized that based on what "some of our politicians (say), you would believe that government creates health innovations and pharmaceutical companies simply profit from them." Yet the reality, he emphasized, "is that it is the private sector that is playing the key role in discovering and developing new therapies."

Hassan went on to say that the exodus of pharmaceutical research from Europe to the United States in recent years was a warning of the dangers of price controls and other government intervention in the high-innovation pharmaceutical industry, an industry that he said was now "one of two pillars of U.S. strength in high technology, along with Silicon Valley." The next president of the United States, said Hassan, should be leading the way "on how best to support one of the leading industries of our country, one that does good things for millions of people -- not arguing over policies to undermine or even to destroy it."

The full article can be found at http://biz.yahoo.com/prnews/040909/nyth179_1.html

Permalink to Comment

3. Jamie McCarthy on September 10, 2004 9:38 AM writes...

Interesting description of the process, thank you. It sounds like you're fighting against the image of drug development just being a matter of picking the low-hanging fruit from government-research trees, and I hope that view isn't too prevalent. Clearly there's a great deal more to it than that.

But serendipity plays a role, too. I live in Kalamazoo, Michigan, where the largest single industry for years has been Upjohn, Pharmacia, and now Pfizer. This town got its economic butt saved by Rogaine, which was originally in testing for something else until someone realized, hey, we're seeing hair growth as a side-effect.

And the conventional wisdom around here is that the development process is more scattershot. It sounds like you're describing an idealized version of the process, but my guess is that many times the target is more of a hunch than a theory, the testing involves dozens (hundreds?) of variant compounds just to see if any one of them is better than the others, and you're always keeping your eyes open for wholly unexpected results, including the next Rogaine.

And while, yes, this process is a marvel of science and engineering, and it is of course very expensive, isn't it true that the pharmaceutical companies are, under the newly relaxed advertising rules, now spending as much money on ads as on R and D?

I'm the one who typed up that Slashdot story (and chose your favorite verb; conveying the importance of news in four to seven words is brutal work sometimes). I'm anxious to hear if I was unfair or incorrect. But it seems undeniable to me that, probably for a mix of reasons including market forces, the pharmaceutical companies are not going to voluntarily disclose results of some, possibly many, clinical trials; that this is bad for science and human knowledge overall; and that if the medical journals can unionize and force this to happen it'll be better for everyone in the long run.

Permalink to Comment

4. Derek Lowe on September 10, 2004 10:50 AM writes...

I agree that it'll be better to have the clinical trial data out where it can be seen (and not least among the consumers will be folks at the other drug companies - there's no way we'd all reveal this to each other unless we were forced to do it.)

The process works as I describe it, but you're right that we need to keep our eyes open. Mostly, unexpected results are unexpectedly bad, though. Having a sudden good effect is pretty rare.

And the variant-compound part is in paragraph 10, which sums up how I spend my days (and years) in a few sentences. The minimum number of compounds that are screened to get a chemical starting point is in the hundreds of thousands, but a figure in the low millions is more common these days.

The minimum number of totally new compounds then made by people like me, from that starting point, is in the high hundreds, but low-to-mid thousands is more common. And long projects have generated more than that. They're made partly for rational hypothesis-testing reasons ("We should make this thing more acidic / longer / greasier / lumpier over in this region") and partly by intuition and wild-eyed guessing ("Wonder what'll happen if we stick this group on - never tried that before.")

Permalink to Comment

5. The Novice Chemist on September 10, 2004 12:49 PM writes...

Hey, Derek: How is the screening of compound libraries (in the low millions) done? Is it done computationally or by some biochemical assay?

Permalink to Comment

6. ehh on September 10, 2004 1:31 PM writes...

But a few questions:

1) Aren't the people conducting Phase 1 - 3 trials are academics, not drug company personnel. You guys are contracting them through CRO's and subcontracting them to write your protocols, but they are still academics emoployed at medical schools whose research capabilities are underwritten at least in part by NIH grants (e.g. JHU is #1 in NIH grants), not to mention their salaries are at least partly paid for by other taxpayer-paid-for sources, state funds, NSF fellowships etc. i.e. you are free-riding in a big way off the government at many steps in the drug development process.

2) If you guys are doing such a great job, why has NIH started its Roadmap program to replicate many of these processes (http://nihroadmap.nih.gov/)?

3)Isn't a large part of the skepticism due to the drug industry's history e.g. the Betty Dong case, the Canadian drugs are evil crap, the refusal to open the books on its spending/mixing of marketing and drug development budgets/me-too drugs? Why should we believe anything you say about the drug industry? It profits from a completely federally-protected market, from NIH coming up with these targets (and bankrolling the other labs involved in their discovery you mentioned), to NSF funding the training of your new hires, to the FDA vouchsafing for your drugs (without which no one would buy them), to the huge tax breaks given to you for R&D and manufacturing, all while PhARMA hides behind dubious proprietary claims and breast-beating about the free market to hide information about matters of human health. I think the problem is bigger than the NIH.

Permalink to Comment

7. Drew on September 10, 2004 1:59 PM writes...

To echo what Chris said, bravo. Now, if we can find way to get this to the masses....
Although, I fear that this might be nearly impossible. One study I read said that 70-80% of adult Americans are unable to put cell, atom and molecule in ascending size order. With a level of scientific illiteracy like that, any description of the drug discovery process is fruaght with peril. It sounds rather trivial and obvious- akin to the PETA camp who seem to think animal testing is unnessesary....

Permalink to Comment

8. Linkmeister on September 10, 2004 4:14 PM writes...

"Aren't the people conducting Phase 1 - 3 trials are academics, not drug company personnel"

No, unless my client is an exception. It's paying a private firm to conduct its trials out of its own pockets, and the contractor is not affiliated or employed by any university.

Permalink to Comment

9. Kram on September 10, 2004 5:09 PM writes...

--This is where the drug industry comes in.--
I would like to point out that even before, there are a whole lot of biotech companies who use state-of-the art technologies (mostly developed by non-academic researchers) to identify new proteins which can serve as interesting compound targets. I work at one such company. Let's not forget that academic research is often really "a shot in the dark", as one commentor put it and that the real flow of target came from the genomics-driven biotechs (and yes, not all that money has been wasted in the genomics years but is now being put to good use by pharma although the results are still far away, because of the lengthy process you describe).
And Derek : thanks for the excellent articles.

Permalink to Comment

10. steve on September 10, 2004 8:12 PM writes...

How many useful therapies are lost because of the novelty requirement in the patent law? In other words, if the space of structures were enriched to include known chemicals that (gasp!) already had been used in humans or animals, would the speed of improve significantly?

Permalink to Comment

11. Maynard Handley on September 11, 2004 1:05 AM writes...

Derek,
You have to realize that the issue is not solely whether or not the drug industry produces new drugs. It is also a host of issues surrounding this. Someone above alluded to the Canadian drugs issue. The imperialism of the Australian Free Trade Agreement
(adopt our way of pricing drugs or else) is an example. The never-ending games of "milk the patent, then change one aspect of the formulation, claim it's new 24hr formula, and start milking the patent again" and so on.

You are trying to focus on one particular thing and claim the drug industry is wonderful because of it. By doing so, you miss the point of the general, inchoate anger against the drug industry. At this point there are various options:
* You can willingly ignore that general anger, continue to fight a very narrow battle, and be surprised that nothing is changing as regards anger against big pharma
* Or you can take on each of these issues that people complain against --- but the problem at that point is that you have no credibility. Sure you are credible wrt the science of the drugs, but you are not party to the financial discussions, the lobbying/bribery of politicians and so on. All you are in a position to do is repeat the same arguments that have been fed to us the public, arguments that specialists in areas like economics and politics frequently find self-serving and wanting.

Permalink to Comment

12. matthew holt on September 11, 2004 12:45 PM writes...

Derek. Nice description as usual, but you'd do well to include two little facts in it. How much money does the NIH spend on basic research and how much does the pharma business spend on it (and you can include development if you like)? I don't have these numbers but I suspect they are closer to each than it would appear from a reader of your article who might think that it's about 90-10 on pharma's side.

However, you correctly state in the previous article that its PhRMA's PR that sucks, and it does. Commenter ehh above suggets that at the least the pharma industry looks like it's hiding something (and we know from numerous settlements that it certainly has been very naughty over on the marketing side of the house). And finally it's called the Pharmaceutical RESEARCH and Maunfacturing Association. Perhaps it should change it's name to the more accurate Pharmaceutical Development and Marketing Association. (TLA pronounced FudMah!) At least that would more acurately reflect what it's doing.

Permalink to Comment

13. Nick Henriquez on September 11, 2004 2:03 PM writes...

Hi Derek,

It's been awile but "plus ca change plus cést la meme chose, nést ce pas?" (no idea how to make the proper symbols here but you get the message).

It seems the discussion about who does what and who should be rewarded (more) has reared it's ugly head again. For those others who read this; I've worked in a very small pharma company but mainly in academia. Derek's post about or "division of labour" is absolutely accurate and to the point.

Where Derek and I will remain in (friendly) disagreement is on the right of companies to any profit they want on anything they produce. Reality on the ground shows that even big pharma is beginning to see that their long-term profits may benefit from a certain amount of "charity" (e.g. "afforable" AIDS drugs fro the 3rd world).

Despite that the aid given is still a drop in the ocean and that the aid supplied is far less than what's been promised, this type of development may obviate all these discussions in future.

As for hiding things, that will quickly become impossible if the editors have their way and they usually do. At least if pharma wants their results published in e.g. the Lancet or Journal of Clinical Investigations ;-).

Not publishing data doesn't only apply to big pharma by the way. Academia is notorious for not publishing "negative" results. This is also to be blamed on editors, but they can be forgiven for not publishing what no-one wants to read (e.g. compound X does not cause hair growth in mice and should not be considered for future therapy).

Despite it's importance I recommend closing this discussion by the way. There's the pros and the cons and just like Bush and Kerry they seem unable to grant the other has a point most of the time (present company excepted of course). We can only hope marketing departments will close their "ministry of truth" and will stick to the facts. That should by and by alter people's perception (e.g. Hey, this stuff actually DOES help me lose weight!). But I guess the pigs will start flying first....

Permalink to Comment

14. ehh on September 11, 2004 2:18 PM writes...

"No, unless my client is an exception. It's paying a private firm to conduct its trials out of its own pockets, and the contractor is not affiliated or employed by any university."

One would think a researcher would know better than to extrapolate from their own situation, especially one they don't seem to understand -- unless your contractor (likely a CRO working with a SMO) has its own stable of MD/PhD's expert in wide areas of medicine (give me a break), it is hiring academics to review protocols, design studies, conduct the trials at centers and write the results up. Small studies may use hospitals and small medical practices but those even usually have their own academic affiliations. This is how clinical trials work, period. See N Engl J Med 2000; 343:1646-1649, Nov 30, 2000 or see a typical center watch trial listing page http://www.centerwatch.com/patient/studies/cat243.html
for atrial fibrillation trials. 3 of 5 are held outright at universities, and of the others, the first one is affiliated with a university hospital and the 2nd one is a one-man shop in Savannah Ga run by a researcher who accumulated his expertise over 25 years as faculty of a medical college.

Every industry loves to make excuses for its free-riding off the government, but at least be honest about it. NIH's budget is $28 billion versus $3.5 billion for the National Science Foundation. Which spending benefits society more?

Permalink to Comment

15. qetzal on September 11, 2004 5:45 PM writes...

You know, ehh, before you go making snarky comments about others, you might want to be sure all your facts are correct, too.

You are correct that academic MDs and clinics play a major role in many clinical trials. That's especially true of large Phase 3 trials for common conditions, where the trial population numbers hundreds or thousands.

But there are also completely private clinics that exist specifically to conduct clinical trials. True, they are typically smaller trials, Phase 1 or 2. And they tend to specialize in certain therapeutic areas. But they are not academic centers in any way and they employ their own medical staff.

So while your basic point was fine, your snotty dismissal of Linkmeister's example was quite off base.

Many of your other points are also not universally true (period).

In any case, are you suggesting there's something wrong with using academic centers and MDs in clinical trials. They do get paid, you know. In my (somewhat limited) experience, it costs $5-10K per patient. That's just what you pay the PI and his institution. It doesn't even cover what you also pay a CRO (or your own people) to monitor the trial, coordinate the lab testing, accumulate & summarize the data, etc., etc. And don't forget making & testing the drug, figuring out how to make and test the drug, handling all the regulatory submissions, and everything else Derek pointed out.

Nevertheless, I can accept that using academics to conduct trials does give industry a "free ride" to some degree. Do you think that's bad? Would you change it somehow? Keep in mind that, to whatever extent NIH supports trials, that is money that industry did not have to spend to get a drug to market, and does not have to recoup from drug sales. Regardless of whether the industry makes fair profits or rapacious ones, I hope you'd agree that making industry pay more to conduct clinical trials will just drive drug prices higher (assuming nothing else changes).

I agree with you that the drug (& biotech) industry has given itself plenty of black eyes, and has at times been quite unethical. I also agree that some of the drug PR is silly. More to the point, Derek has consistently called his own industry to task on such points, in complete agreement with some of your issues. So here you have an industry insider whose actually your potential ally (to a degree), and your response is to be snide and dismissive.

All you've done is demonstrated that it is YOU who does not deserve to be listened to.

Permalink to Comment

16. John Thacker on September 12, 2004 4:08 AM writes...

qetzal--

I hope you'd agree that making industry pay more to conduct clinical trials will just drive drug prices higher (assuming nothing else changes).

I think, since drug prices are generally set at maximum profit prices while under patent, the more likely result is that more drugs become unprofitable, and fewer drugs are researched. The ones that still exist probably have the same, profit-maximizing price. They could be higher with less competition, though I suspect that me-too drugs would still be popular as they are somewhat more likely to be successful and hence profitable. So price is harder to say; almost surely, though, there would be fewer drugs discovered.

It's always a hard question, I grant, when it comes to deciding between cheaper drugs now to save money and lives, and more drugs discovered for the future to save money and lives then. A lot of people are willing to sacrifice the future for the present, and in specific cases it's hard to blame them.

Permalink to Comment

17. ehh on September 13, 2004 1:12 PM writes...

"Talk sense to a fool and he calls you foolish."
-- Euripdes

quetzal -- after suggesting I'm mostly right and that you agree with one of my main points, you dismiss my comment as incorrect. Cognitive dissonance much? You also misunderstand, disingenuously or not, my points. There is nothing wrong with Pharma leaning on NIH-underwritten academic centers to run most of its trials (especially the very expensive Phase 3 ones.) But it's incorrect to say that drug discovery is divorced from academia, which was the point of the discussion that started these comments. The entire drug industry, from start to finish is enmeshed in government subsidies, even in the drug discovery process. Would we have more, cheaper or better drugs under a different system? Beats me. But let's at least be honest about the system we have.

Derek is refreshingly open about PhARMA's follies, no doubt. I just wanted him to address the full picture, that's all. He is too open-minded to be troubled by a little snarkiness, I'd say.

And no one has suggested my minor point about the NIH roadmap or the industry's larger problems are incorrect.

-- But I should note that NSF's full budget is $4.7 billion, not $3.5 billion, which is the research budget for the agency. Also since some people asked, "In 2003, PhRMA member companies invested an estimated $33.2 billion on research to develop new treatments for diseases," according to PhRMA.

Permalink to Comment

18. james on September 13, 2004 2:49 PM writes...

The bigger PR nightmare for Pharma is the disparity in pricing. Voters want to know why a drug sold in Canada can cost 20x less than the same drug (same manufacturer) sold in the US. The second thing voters want to know is why the generic version sold in the US is so much cheaper than the name brand version. The fact the the US a free market for drug sales will not fly. The second issue could be successfully passed off as including cost of initial development. However, Pharma keeps shooting itself in the foot by the under handed efforts to extend patent life. Through the law suite loophole automatically granting a 2 year stay or collusion with the generic manufactor.

Permalink to Comment

19. jg on September 13, 2004 8:03 PM writes...

The basic fact about drug company economics is this: for decades they have been making excess profits; that is, achieving a return on investment higher than the norm for all industries. In a competitive market this is supposed to be impossible, and implies that costs for the user are higher and less is comsumed than optimal.

This fact is the driver for anger against Pharma. Of course, their secretiveness about anything concerning costs of research vs. marketing and the "me too" drug phenomon also provide fuel, as to the patent extension games described above.

None of which disagrees with anything Derek said, or disparages his and his colleagues' efforts. But from the standpoint of evaluating, e.g. the social utility of various intellectual schemes his description is largely beside the point.

Permalink to Comment

20. biff on September 14, 2004 9:05 AM writes...

The basic fact about drug company economics is this: for decades they have been making excess profits; that is, achieving a return on investment higher than the norm for all industries. In a competitive market this is supposed to be impossible

Sorry, but that statement is just blindingly ignorant of basic economics. Each industry has its own "norm" for return on investment based on things like the risk of the business, the difficulty of business, the capital / talent required to run the business, the demand for product, etc. This is why the potential ROI is higher in fields like high technology, entertainment, etc. than in the grocery and newspaper businesses. If the potential ROI wasn't quite high in Pharma, there would be no reason for investors to invest in such a high risk endeavor! They'd be better off putting their money in, well, supermarkets!

Permalink to Comment

21. Tom on September 14, 2004 1:47 PM writes...

"In other words, if the space of structures were enriched to include known chemicals that (gasp!) already had been used in humans or animals, would the speed of improve significantly?"

Isn't this done already with alternate indications? CNS pharmaceuticals like gabapentin are the paradigmatic examples of indication-creep. (It's anti-seizure! It's a chronic pain reliever! It's an anxiety therapeutic!)

Permalink to Comment

22. MQ on September 14, 2004 4:41 PM writes...

JG above has the right way of thinking about this. None of this justifies drug companies earning massive excess profits on the back of patients. You need to get at the question of whether drug company profits are excessive. Almost all industries do real work for society, but only pharma gets completely government-protected markets for critical goods with low elasticities of demand, which is a setup for enormous monopoly profits. This is really what the debate is about.

And even in your account publicly funded academia/NIH is an important part of the drug development process. But consumers get no benefit or break in drug pricing in exchange for that role. Drug companies have complete freedom to set the monopoly price.

Permalink to Comment

23. John Thacker on September 14, 2004 11:52 PM writes...

Nope, MQ, you don't understand basic economics well. A little knowledge is a dangerous thing. You're completely wrong in this sentence:

But consumers get no benefit or break in drug pricing in exchange for that role.

Consumers do get a benefit. They get MORE DRUGS BEING DISCOVERED AND RESEARCHED. This sort of research rarely lower drug prices, that is true. Prices are generally set at profit maximizing levels regardless. However, such subsidies and research DOES end up making some drugs profitable to research and produce that otherwise would not be produced. THAT'S the difference; it makes the market larger. By making all drugs more profitable, companies can now make money on producing some drugs that otherwise wouldn't be worth it. Yes, drugs that would be profitable anyway become more profitable, and yes, they have monopoly price power. Completely true. However, the consumer payoff is the new drugs that otherwise wouldn't exist. Even with monopoly price power, some drugs just aren't worth it to produce-- e.g., the number of people with the disease might be too low to justify research costs. With enforced low prices, even more drugs become too expensive to produce.

However, while people can easily see and notice the high price of existing drugs, and notice the benefits of lowering the price, but can't measure the deleterious effects of not inventing more drugs easily (how do you measure something that never was?), this gets discounted.

Save lives now or save lives later, such a choice to make.

Permalink to Comment

24. MQ on September 15, 2004 12:17 PM writes...

Interesting that blog comments so often begin with an obligatory personal attack on the person being responded to. I can guarantee you that I understand basic economics well, John -- I think I've probably forgotten more about it than you know.

Anyway, my statement was that consumers get no benefit or break in drug pricing from the NIH/university subsidies they pay for drug research. Given that a drug is invented, this is a completely true statement -- consumers get no price benefits in return for their contribution to the fixed cost of developing a new drug. (The subsidy is basically to fixed costs, so it doesn't lower the company MC curve). This is different than saying that the public role in drug development and research brings no public benefit at all -- clearly it does.

From what i can make out of your response, you appear to have thought I was saying that consumers got no benefit from the (government supported) patent system. This is a much more complex issue and was certainly not what I was saying, I was referring to the NIH/academia role in drug development. Given that we have a private sector drug industry, we need a regulatory system to ensure that SOMEONE pays the fixed costs of drug development work performed by the drug industry. Without some such regulatory system we would have very few drugs produced. We've chosen to use a patent system, which tends to assign those fixed costs to patients (and their insurance companies), and may also give the drug companies a monopoly profit (in excess of normal rates of return) on top of their fixed development costs. Using the patent system to guarantee fixed cost reimbursement plus profits is clearly better than using NO system. But the question of whether the patent system is the BEST regulatory system we could use to guarantee reasonable returns to drug producers is still very much open. Alternatives are more tightly regulating private pharma pricing (which does not mean eliminating a normal rate of return), nationalizing more of the system (which could go along with generous salaries for successful labs), etc. Each alternative has its pluses and minuses. But our current patent-based system has it minuses as well...notably that the financial burden of drug development is not equitably distributed and some companies may be able to use government protections to earn supranormal profits.

Permalink to Comment

25. John Thacker on September 16, 2004 10:43 AM writes...

As I said, MQ, consumers do indeed not get price benefits in return for their contribution to the fixed cost of developing a new drug. What they do get, certainly, is more drugs researched because more drugs are profitable to research. This is completely separate from the benefits (and costs) of the patent system-- although both involve the same sort of tradeoff.


The patent system does not in any way "assign the fixed costs to the patients." That's, IMO, a misleading way to put it. Rather, it lets the drug companies charge a monopoly price for their product. IF (the company expects that) the total profit the drug company can make from the drug exceeds the fixed costs, then they'll produce it. (Of course, in reality there are various determinations of the likelihood of success of developing a drug, since it's rarely simple. But the basic equation holds; if expected profit is higher than expected costs, then an effort will be made to develop it.)


The patent system does not "tend to assign those fixed costs to patients" any more than not having a patent system does. Companies do not long stay in business unless their fixed costs are covered by their profits. Rather, the patent system increases the expected profits of the drug company, allowing them to exceed the fixed costs in many more instances, thus causing more pharmaceuticals to be profitable to research.

Any sort of "cost plus" system to regulate prices to get a "normal rate of return" would have a huge problem in figuring out how to allow for all the money spent on research that doesn't pan out. Most systems I can think of either wouldn't properly reimburse for such (and thus would highly discourage research, especially innovative, high-risk high-reward research), or would create perverse incentives in the wrong direction, such as incentives to game the system and perform useless research, since it would still be reimbursed.

Ranting about "supranormal monopoly profit" and "normal rates of return" is one of the signs of failing to understand economics. Monopoly profit occurs in any non-commodity, non-perfect competition situation-- which also means any situation where there is technological advance, product differentiation, and new products being researched and introduced. It is the race for monopoly profits that drives research and technological advances-- the drive to capture monopoly profits for the limited amount of time until a competitor can catch up. If you legislate away monopoly profits, you will legislate away the incentives for innovation. The recourse, as you note, is nationalizing. (In some cases the tradeoff may be worth it, or at least considering.)

MQ, in addition, while drugs may have low elasiticity of demand as a whole, there are substitutes for most drugs. (Note that "me-too" drugs are helpful here.) Given the existence of substitutes, the monopoly profits are not nearly as large as you claim.

At least you aren't making the ridiculous anti-advertising claims that so many do.

Permalink to Comment

26. qetzal on September 16, 2004 10:54 AM writes...

Nice post, MQ. Oops, sorry! I was supposed to attack you, right? ;-)

I completely agree that the key question is not, "What is right or wrong with the current system?" It is, "What is the best system we could have?"

To some extent, can we not benchmark our system by looking at successful drug development across countries? I can think of a lot of reasons why that's an imperfect measure, but perhaps it is still informative. My impression is that drug development is more successful in the US than anywhere else. It would be interesting to know if that impression is supported by real data. If so, what could we reasonably conclude from that?

[Qualifiers: Note that I'm talking about drug development in the US. I'm perfectly aware that much of the drug development done outside the US is still being done for the US, since profit potential is usually much higher here. And if you're developing for the US, but you're not in the US, there are certainly all kinds of potential barriers (laws, culture, language, geographical distance, etc.) that a US company wouldn't face. We'd also have to recognize that most of the knowledge generated by NIH-funded work becomes available worldwide through publication (which helps people in other countries), but the patent rights are not freely available (which hurts), etc., etc. It's definitely something that would need very careful analysis before making any conclusions.]

Permalink to Comment

27. MQ on September 17, 2004 1:47 AM writes...

Well, yes, neoclassical economics ignores the necessity to create incentives for technological change, and supranormal profits at some point and some level are necessary if you want to create those incentives through the profit system. You've got me there, although I would always have agreed on that conceptually; I wasn't meaning to imply incentives aren't necessary. But perhaps you could also share the source of your certainty that a 20-year government guaranteed monopoly to private drug developers provides the optimal level of those incentives for the money, and the large-scale transfer of wealth from sick patients to pharma shareholders is the fairest way to provide them.


There are lots of ways to create incentives, ranging from regulated prices that still maintain sizable profits (which one could get through a shortened patent period) to things like direct government financial awards to developers of new or popular drugs. This always struck me as a good way to reimburse fixed costs. Granted, neoclassical economics doesn't pay enough attention to incentives, but an Austrian focus on profit incentives alone ignores the very real and very large economic inefficiencies created by not providing cheap-to-manufacture drugs to users who desparately need them but can't afford to pay. Not to mention that providing incentives solely through the profit system ignores the potentially large economic inefficiencies created by the lack of alignment between medical need and wealth.


Just saying "incentives are important" is no more compelling as an argument for the current system than the statement that "drug companies make money" is as a point against it. You have to argue for why the current system is the best one could come up with. You correctly point out problems with cost-plus systems, but there are many ways to provide more limited rewards than we currently do that are still contingent on the success of a drug.


And there is already plenty of gaming under the current system. Me-too drugs, while they do reduce the pricing power of competitor drugs, are a cost of the current system not an argument for them. More direct regulation of prices could have the same effect without spending lots of resources looking for a biochemically new way to do the same thing for patients. (Although sometimes these drugs are valuable as incremental improvements over the old ones).

Permalink to Comment

28. hope on September 17, 2004 3:24 PM writes...

Derek, I enjoyed your article.

You and your commenters are all much more knowledgeable than I am about this stuff. But I'll comment anyway. :) It seems to me that some balance should be struck between maintaining the potential profits necessary to encourage R&D and ensuring that people who need a drug can get it. After all, it isn't in anyone's interests - pharma industry's included - for drug prices to be so high that they negatively impact necessary utilization. Maybe the industry doesn't perceive drug prices as too high from that standpoint?

Permalink to Comment


EMAIL THIS ENTRY TO A FRIEND

Email this entry to:

Your email address:

Message (optional):




RELATED ENTRIES
Conference in Basel
Messed-Up Clinical Studies: A First-Hand Report
Pharma and Ebola
Lilly Steps In for AstraZeneca's Secretase Inhibitor
Update on Alnylam (And the Direction of Things to Come)
There Must Have Been Multiple Chances to Catch This
Weirdly, Tramadol Is Not a Natural Product After All
Thiola, Retrophin, Martin Shkrell, Reddit, and More