Are Drug Companies Faking an Innovation Crisis? Uh, No.

By Guest Blogger | August 16, 2012 12:39 pm


Derek Lowe is a medicinal chemist who has worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer’s, diabetes, osteoporosis, and other diseases. He has been writing about drug discovery at In the Pipeline, where this post originally appeared, for more than ten years.

The British Medical Journal says that the “widely touted innovation crisis in pharmaceuticals is a myth.” The British Medical Journal is wrong.

There, that’s about as direct as I can make it. But allow me to go into more detail, because that’s not the the only thing they’re wrong about. This is a new article entitled “Pharmaceutical research and development: what do we get for all that money?”, and it’s by Joel Lexchin (York University) and Donald Light of UMDNJ. And that last name should be enough to tell you where this is all coming from, because Prof. Light is the man who’s publicly attached his name to an estimate that developing a new drug costs about $43 million dollars.

I’m generally careful, when I bring up that figure around people who actually develop drugs, not to do so when they’re in the middle of drinking coffee or working with anything fragile, because it always provokes startled expressions and sudden laughter. These posts go into some detail about how ludicrous that number is, but for now, I’ll just note that it’s hard to see how anyone who seriously advances that estimate can be taken seriously. But here we are again.

Light and Lexchin’s article makes much of Bernard Munos’ work (which we talked about here), which shows a relatively constant rate of new drug discovery. They should go back and look at his graph, because they might notice that the slope of the line in recent years has not kept up with the historical rate. And they completely leave out one of the other key points that Munos makes: that even if the rate of discovery had remained linear, the costs associated with it sure as hell haven’t.

No, it’s all a conspiracy:

“Meanwhile, telling “innovation crisis” stories to politicians and the press serves as a ploy, a strategy to attract a range of government protections from free market, generic competition.”

Ah, that must be why the industry has laid off thousands and thousands of people over the last few years: it’s all a ploy to gain sympathy. We tell everyone else how hard it is to discover drugs, but when we’re sure that there are no reporters or politicians around, we high-five each other at how successful our deception has been. Because that’s our secret, according to Light and Lexchin. It’s apparently not any harder to find something new and worthwhile, but we’d rather just sit on our rears and crank out “me-too” medications, drugs similar to those that already exist, for the big bucks:

“This is the real innovation crisis: pharmaceutical research and development turns out mostly minor variations on existing drugs, and most new drugs are not superior on clinical measures. Although a steady stream of significantly superior drugs enlarges the medicine chest from which millions benefit, medicines have also produced an epidemic of serious adverse reactions that have added to national healthcare costs.”

So let me get this straight: according to these folks, we mostly just make “minor variations,” but the few really new drugs that come out aren’t so great either, because of their “epidemic” of serious side effects. Let me advance an alternate set of explanations, one that I call, for lack of a better word, “reality.” For one thing, “me-too” drugs are not identical, and their benefits are often overlooked by people who do not understand medicine. There are overcrowded therapeutic areas, but they’re not common. The reason that some new drugs make only small advances on existing therapies is not because we like it that way, and it’s especially not because we planned it that way. This happens because we try to make big advances, and we fail. Then we take what we can get.

No therapeutic area illustrates this better than oncology. Every new target in that field has come in with high hopes that this time we’ll have something that really does the job. Angiogenesis inhibitors. Kinase inhibitors. Cell cycle disruptors. Microtubules, proteosomes, apoptosis, DNA repair, metabolic disruption of the Warburg effect. It goes on and on and on, and you know what? None of them work as well as we want them to. We take them into the clinic, give them to terrified people who have little hope left, and we watch as we provide with them, what? A few months of extra life? Was that what we were shooting for all along, do we grin and shake each others’ hands when the results come in? “Another incremental advance! Rock and roll!”

Of course not. We’re disappointed, and we’re pissed off. But we don’t know enough about cancer (yet) to do better, and cancer turns out to be a very hard condition to treat. It should also be noted that the financial incentives are there to discover something that really does pull people back from the edge of the grave, so you’d think that we money-grubbing, public-deceiving, expense-padding mercenaries might be attracted by that prospect. Apparently not.


The same goes for Alzheimer’s disease. Just how much money has the industry spent over the last quarter of a century on Alzheimer’s? I worked on it twenty years ago, and God knows that never came to anything. Look at the steady march, march, march of failure in the clinic—and keep in mind that these failures tend to come late in the game, during Phase III, and if you suggest to anyone in the business that you can run an Alzheimer’s Phase III program and bring the whole thing in for $43 million dollars, you’ll be invited to stop wasting everyone’s time. Bapineuzumab’s trials have surely cost several times that, and Pfizer/Johnson & Johnson are still pressing on with it. And before that you had Elan working on active immunization for Alzheimer’s, which is still going on, and you have Eli Lilly and Co’s antibody trials, which are still going on, and Genentech’s (which are still going on). No one has high hopes for any of these, but we’re still burning piles of money to try to find something. And what about the secretase inhibitors? How much time and effort has gone into beta- and gamma-secretase? What did the folks at Lilly think when they took their inhibitor way into Phase III only to find out that it made Alzheimer’s slightly worse instead of helping anyone? Didn’t they realize that professors Light and Lexchin were on to them? That they’d seen through the veil and figured out the real strategy of making tiny improvements on the existing drugs that attack the causes of Alzheimer’s? What existing drugs to target the causes of Alzheimer’s are they talking about?

Honestly, I have trouble writing about this sort of thing, because I get too furious to be coherent. I’ve been doing this sort of work since 1989, and I have spent the great majority of my time working on diseases for which no good therapies existed. The rest of the time has been spent on new mechanisms, new classes of drugs that should (or should have) worked differently than the existing therapies. I cannot recall a time when I have worked on a real “me-too” drug of the sort of that Light and Lexchin seem to think the industry spends all its time on.

That’s because of yet another factor they have not considered: simultaneous development. Take a look at that paragraph above, where I mentioned all those Alzheimer’s therapies. Let’s be wildly, crazily optimistic and pretend that bapineuzumab manages to eke out some sort of efficacy against Alzheimer’s (which, by the way, would put it right into that “no real medical advance” category that Light and Lexchin make so much of). And let’s throw caution out the third-floor window and pretend that Lilly’s solanezumab actually does something, too. Not much—there’s a limit to how optimistic a person can be without pharmacological assistance—but something, some actual efficacy. Now here’s what you have to remember: according to people like the authors of this article, whichever of these antibodies that makes it though second is a “me-too” drug that offers only an incremental advance, if anything. Even though all this Alzheimer’s work was started on a risk basis, in several different companies, with different antibodies developed in different ways, with no clue as to who (if anyone) might come out on top.

All right, now we get to another topic that articles like this latest one are simply not complete without. That’s right, say it together: “Drug companies spend a lot more on marketing than they do on research!” Let’s ignore, for the sake of argument, the large number of smaller companies that spend all of their money on R&D and none on marketing, because they have nothing to market yet. Let’s even ignore the fact that over the years, the percentage of money being spent on drug R&D has actually been going up. No, let’s instead go over this in a way that even professors at UMDNJ and York can understand it:

Company X spends, let’s say, $10 a year on research. (We’re lopping off a lot of zeros to make this easier). It has no revenues from selling drugs yet, and is burning through its cash while it tries to get its first on onto the market. It succeeds, and the new drug will bring in $100 dollars a year for the first two or three years, before the competition catches up with some of the incremental me-toos that everyone will switch to for mysterious reasons that apparently have nothing to do with anything working better. But I digress; let’s get back to the key point. That $100 a year figure assumes that the company spends $30 a year on marketing (advertising, promotion, patient awareness, brand-building, all that stuff). If the company does not spend all that time and effort, the new drug will only bring in $60 a year, but that’s pure profit. (We’re going to ignore all the other costs, assuming that they’re the same between the two cases).

So the company can bring in $60 dollars a year by doing no promotion, or it can bring in $70 a year after accounting for the expenses of marketing. The company will, of course, choose the latter. “But,” you’re saying, “what if all that marketing expense doesn’t raise sales from $60 up to $100 a year?” Ah, then you are doing it wrong. The whole point, the raison d’etre of the marketing department is to bring in more money than they are spending. Marketing deals with the profitable side of the business; their job is to maximize those profits. If they spend more than those extra profits, well, it’s time to fire them, isn’t it?

R&D, on the other hand, is not the profitable side of the business. Far from it. We are black holes of finance: huge sums of money spiral in beyond our event horizons, emitting piteous cries and futile streams of braking radiation, and are never seen again. The point is, these are totally different parts of the company, doing totally different things. Complaining that the marketing budget is bigger than the R&D budget is like complaining that a car’s passenger compartment is bigger than its gas tank, or that a ship’s sail is bigger than its rudder.

OK, I’ve spend about enough time on this for one morning; I feel like I need a shower. Let’s get on to the part where Light and Lexchin recommend what we should all be doing instead:

What can be done to change the business model of the pharmaceutical industry to focus on more cost effective, safer medicines? The first step should be to stop approving so many new drugs of little therapeutic value. . .We should also fully fund the EMA and other regulatory agencies with public funds, rather than relying on industry generated user fees, to end industry’s capture of its regulator. Finally, we should consider new ways of rewarding innovation directly, such as through the large cash prizes envisioned in US Senate Bill 1137, rather than through the high prices generated by patent protection. The bill proposes the collection of several billion dollars a year from all federal and non-federal health reimbursement and insurance programmes, and a committee would award prizes in proportion to how well new drugs fulfilled unmet clinical needs and constituted real therapeutic gains. Without patents new drugs are immediately open to generic competition, lowering prices, while at the same time innovators are rewarded quickly to innovate again. This approach would save countries billions in healthcare costs and produce real gains in people’s health.

One problem I have with this is that the health insurance industry would probably object to having “several billion dollars a year” collected from it. And that “several” would not mean “two or three”, for sure. But even if we extract that cash somehow—an extraction that would surely raise health insurance costs as it got passed along—we now find ourselves depending on a committee that will determine the worth of each new drug. Will these people determine that when the drug is approved, or will they need to wait a few years to see how it does in the real world? If the drug under- or overperforms, does the reward get adjusted accordingly? How, exactly, do we decide how much a diabetes drug is worth compared to one for multiple sclerosis, or TB? What about a drug that doesn’t help many people, but helps them tremendously, versus a drug that’s taken by a lot of people, but has only milder improvements for them? What if a drug is worth a lot more to people in one demographic versus another? And what happens as various advocacy groups lobby to get their diseases moved further up the list of important ones that deserve higher prizes and more incentives?

These will have to be some very, very wise and prudent people on this committee. You certainly wouldn’t want anyone who’s ever been involved with the drug industry on there, no indeed. And you wouldn’t want any politicians—why, they might use that influential position to do who knows what. No, you’d want honest, intelligent, reliable people, who know a tremendous amount about medical care and pharmaceuticals, but have no financial or personal interests involved. I’m sure there are plenty of them out there, somewhere. And when we find them, why stop with drugs? Why not set up committees to determine the true worth of the other vital things that people in this country need each day—food, transportation, consumer goods? Surely this model can be extended; it all sounds so rational. I doubt if anything like it has ever been tried before, and it’s certainly a lot better than the grubby business of deciding prices and values based on what people will pay for things (what do they know, anyway, compared to a panel of dispassionate experts?)

Enough. What we have here is someone’s fantasy about how drug discovery works, not the reality. Profs. Light and Lexchin don’t seem to have noticed that the pharma industry has been laying off thousands of people in recent years, or that the stocks of most of the publicly traded companies haven’t been very strong investments. If it’s as easy (and as cheap) to discover new drugs as they claim, we should be fighting off the investors, but where are they? Drug companies are certainly not consistent angels, but neither are they devils—and the last thing they are is lazy and complacent devils, at that.

Pill images via Shutterstock

CATEGORIZED UNDER: Health & Medicine, Top Posts
  • Simon


    If I spend $10/year on R to bring in $60/year, and then spend $30/year on M to add another $40 a year income, I’ll have spent $40 to get $100.

    Could I not just spend $40 on R and get $240 back?

    Your example only works if the rate of return on M is larger than that of R. Which is probably somebody else’s point.

    Of course, if you’re competing in a market of similar products, the ROI of R & M will not be simple multipliers; there will be a maximum total ROI somewhere with a particular mixture of R & M. This point is unlikely to coincide with the point of maximum public good – but that’s capitalism for you, I expect the alternatives are worse.

  • Pingback: R&D | Kristine Smith()

  • Plato

    As somebody who is extremely frustrated with the pharmaceutical industry, I found this post very enlightening. I completely agree with the last paragraph, which outlines a system of government similar to Plato’s aristocracy in the republic. I’m afraid that designing drugs to cure disease is bad science to begin with. Although we got lucky with a few drugs, I don’t think there will ever be a drug cure for cancer, or diabetes, or mental illness, etc. But drugs do help. The X factor is consciousness. Consciosiness is not factored into science or medicine, certainly not disease. Why do some people spomtaneously remit from a disease? Why is a positive attitude so important? Some of the most effective drugs for certain ailments might also be some of the most conscious altering. But ooohhhhhh that’s a no no. Don’t make any of those drugs!!! Will they cure disease by themselves? No. But they might be able to be used as tools in the right hands, like a surgeon’s toolbox. Until we start doing that, youre just wasting money and letting people suffer.

  • sb

    If drug companies were more transparent about their finances, these ‘assumptions’ wouldn’t need to be made. But they’ve always refused to.

  • John Appleby

    Interesting blog.

    Re penultimate para, in England of course, we have tried the rational response to the value question with a health technology assessment process overseen by NICE for over a decade. And it seems, in general, to work well – especially given the difficult trade offs and judgements involved (as you note) and is generally seen as better (for individuals and society) than simply relying on willingness to pay to decide allocation and production activity.

  • Robert (Jamie) Munro

    The marketing people spend $30 to get another $40 of sales. Let’s call that an extra 40 people treated with the drug, $1 per person.

    Who are those people? If the drug is best for them, then their doctors are incompetent for not knowing the range of drugs properly before the marketing. If the drug is not best for them, then the doctors are incompetent for being biased by marketing departments into giving the wrong drug.

    IMHO, drug company marketing is not the right way to be telling doctors what to prescribe. They should be using some sort of independently created reviews of drugs, produced by medical journals, doctors associations, or organisations like NICE in the UK.

  • Pingback: Assorted links()

  • Kaviani

    The problem with this industry is very much the same with others. Execs are demanding profit and refusing to pay any more than they did before. Increase productivity w/o costs. That amounts to serious layoffs, halfass outsourcing, and general devolution simply to increase bottom lines. Innovation is secondary to profit, because profits can be “made” without it.

    The vigorous lobbying (leading to dubious FDA fasttracking *in some instances*) and marketing around certain companies coupled with good ol’ American paranoia work in tandem to create the distrust of pharmaceuticals. Just check the anti-vax freaks if you want a bellweather.

  • AJ

    Gee, I wonder what pharmaceutical companies did before they were allowed to advertise prescription drugs directly to the general public in the US. Oh that’s right, they DIDN’T spend billions on advertisement campaigns and because of that, prescription drugs cost less. Now we have drugs like Humira which cost $15,000 a year – or even more! – in the US.

    Also, why is it that a drug like Humira costs so much more in the US than in ANY other country? Go to Finland, Canada, Germany, UK, etc. and Humira is considerably less expensive. There’s only a few possibilities as to why – and none of them are good news for US patients.

  • Lon Jones

    When I started practicing a new drug cost about a million, then three, then seven, and now 43. WOW! That escalation beats inflation all to pieces. Likely is a close reflection, however, of the growth in our understanding of human complexity. If regulating agencies demand that every ‘i’ be dotted and every ‘t’ crossed in demonstrating safety and efficacy and the blossoming of those elements continues with every new technology we do indeed have a problem. Analyzing complex systems is what this game is all about, and it doesn’t work; there are too many connections. Complex systems are best predicted by computer modeling; weather being the best example. Lots of luck modeling human complexity, however; throw in the characteristic of adaptability and even good computer models don’t have a chance. The solution we discuss in our book, The Boids and the Bees, is to look only at safety for drugs–which allows us to use the placebo effect. And we also need to honor Plato’s POV and allow more drugs used wisely and well.

  • orthodoc

    “If drug companies were more transparent about their finances, these ‘assumptions’ wouldn’t need to be made. But they’ve always refused to.”

    Right. No way any of these companies would ever, you know, put their financial data on line where the public could look at it.

    Whoops, I guess they do. Here you go. Have fun.

  • Uncle Al

    Human Resources (Personnel) specifically does not hire the kind of mutant smart, obnoxiously creative, Asperger-Google kind of guy who could save you. HR hires mediocrities acceptable as drinking buddies. Management exists to kill the future, for the only trusted employee is one whose sole marketable asset is loyalty. At every level of education, rather than foster brilliance we allocate for its suppression. The behaviorally and especially the intellectually aggressive male must be destroyed as national social policy.

    Compassion-building exercise: How many bad hires are needed to sum to one Manhattan Project? There’s your problem – deformed decisions, administrative soft despotism; a tyranny of immersive falsehoods. Physical reality only has one vote. It’s a whopper.

  • Ken Rubenstein

    I’ve never worked in pharma, but as an analyst/journalist I’ve interviewed scores of pharma researchers and others in the drug discovery world, and it’s TOTALLY obvious to me that the innovation deficit is very real. There several reasons for it, but the main one is that pharmas, grown huge by mergers (which they did to compensate for the innovation deficit), felt they had to tackle common, chronic diseases that are poorly understood in terms that are translatable to drug discovery. Basic research on genomes, cells, and tissues probably needs another decade or more before pharmas have enough info to address these illnesses in rational and predictable ways. Until then, the marketing types will be forced to struggle for company survival (not assured) by hook or crook. Caveat emptor!!

    You see, in the capitalist system (not that I’m in love with all aspects of it), companies need to make every effort to generate returns for their shareholders or the class action lawyers get busy doing their thing.

  • AsiaFan

    I am only a layperson with regard to this industry, but my father was in this industry and I saw how difficult it can be. Thank you for posting this article.

  • Geack

    @3. Plato,
    “Consciousness” can cure diseases, therefore consciousness-altering drugs might cure more disease? C’mon… your basic thesis is contradicted by all of human history, in which people who really really believe they can and should live have died from disease. Some people experience remission for the same reasons only some people in a given place catch the flu – bodies are massively complicated and constantly changing, disease vectors vary in dose, potency, and susceptibility, and we have built-in systems that try to fix problems. Sometimes everything lines up right, even unexpectedly. It’s not due to their superior “consciousness”. Positive attitude is completely irrelevant beyond maintaining the basic will to live that keeps you eating enough to give your body a chance. Whatever mind-over-matter nonsense you’re teasing in your post, please don’t confuse it with medicine.

  • Geack

    @10. Lon Jones,
    “…which allows us to use the placebo effect.” Placebos don’t DO anything. The placebo effect is the tendency of some people to occassionally be briefly less bothered by their symptoms after experiencing something they believe might help. Since you get placebo effects both from treatments known to work and from treatments not known to work, it’s unethical to try to invoke a placebo effect from a treatment not know to work. You’re basically conning the patient into thinking something’s changed for a little while without even attempting to address the cause of the symptom.


The Crux

A collection of bright and big ideas about timely and important science from a community of experts.

See More


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar