Skip to main content

Academia (vs. Industry)

Where Drugs Come From: A Comprehensive Look

This is a solid article by Jeffrey Flier (open-access in the Journal of Clinical Investigation) on the roles of academia and industry in drug discovery. Which is a topic that refuses to go away. I am prepared to swear that before starting this blog I had no idea of how many people are convinced that the drug industry does little or no research, and that academic labs discover all the drugs.

I will gladly send anyone unfamiliar with the real situation to this new article, because I think it does an excellent job of laying out just who does what in this business. Short answer: it’s a team effort, of course, and we all do different (and valuable) things. As Flier puts it, “Most fundamental biologic insights have resulted from work by academic scientists. . .“, but the view that all industry does is monetize academic discoveries “reflects a limited understanding of the work carried out by the pharmaceutical and biotech industries“. I could not agree more. When someone says that industry doesn’t do research, etc., I just want to ask them what I’ve been up to for. . .well, thirty years this fall. I hope that this article proves especially valuable because it’s coming from someone in academia – my own arguments on the subject, while vigorous and heartfelt, are surely taken as tendentious special pleading by the people whose views I’d most like to change. That link, though, will take you to over ten years of said arguments, for what that’s worth.

For even more chapter and verse, here’s the Congressional Research Service (PDF) in a recently updated report on the National Institutes of Health. I especially recommend the section starting on numbered page 37 on the balance between academia and industry, and a few pages later on publicly funded research and drug development:

Since over 50% of NIH funding supports basic research, NIH funded research is, to a greater extent, indirectly involved—by generating scientific knowledge and innovations that aid in pharmaceutical development. For example, important basic advances in research, such as recombinant DNA, can lead to the development of whole new classes of drugs. NIH also supports the education and training of biomedical scientists, some of whom then work for the pharmaceutical industry. It is therefore difficult to quantify and assign credit for the role of NIH funding in the development of a given drug.

The CRS goes on to reference several previous attempts to quantify how many drugs can be directly attributed to public-sector research, such as the 2011 study that estimated the 1990-2007 share at about 9%. The overall picture is identical to Flier’s paper above (note: that’s because it’s right):

Rather than directly leading to a new drug, NIH-funded researchers are more often indirectly involved in drug development by producing scientific research and innovations that contribute to the knowledge base and available methods for the pharmaceutical industry.

Exactly. Here’s another such study that appeared last year from a group at the Center for the Integration of Science and Industry at Bentley University, which found (from one direction) that NIH-funded basic research contributed in some way to every single one of the 210 FDA-approved drugs from 2010-2016, but (from the other direction) that “>90% of this funding represents basic research related to the biological targets for drug action rather than the drugs themselves“. Both of those conclusions sound correct from my experience as well. The basic research funded by the NIH and others is absolutely crucial to finding drugs. But the applied research in industry is absolutely crucial, too. Both of these things can be true at the same time, and both of them are.

Here’s another excellent look at the issue, from a professor at Stanford’s law school (Lisa Larrimore Ouellette). She specializes in patent law as it applies to translational research, and her conclusions are completely in line with what’s described above. Her post also addresses the returns on investment to the public seen from NIH research, whether everyone is “paying twice” or not when such research is commercialized, and what the various incentives really are for publicly- and privately-funded drug research. I strongly recommend anyone interested in these issues read it; it’s exceptionally detailed and clear-headed. Part of the frustration in talking about such issues, in fact, is dealing with people who have not thought about these issues in any detail, who may not realize that many of these issues even exist, but are nonetheless passionately convinced that they have grasped all they need to know.

I think that some of the confusion about where drugs come from originates from people who don’t understand how research works in general. Some of them appear to have a mental landscape consisting mostly of thieves and victims, with everything and everyone being assigned as one or the other. But if Research Group A, back in the 1980s, identifies a particular enzyme in cells, and Group B in the 2010s uses genetic variants of it in the human population to propose that it’s unexpectedly important in some disease, and Group C screens their company’s compound collection and develops a small-molecule drug against it and takes it into clinical trials after many twists and turns, here’s the thing: none of these people are ripping any of the others off. They’re all doing science, and this is how science works. Group A was using techniques first invented by others and refined by still more people, and Group B were using sequencing technologies that they didn’t invent, either, and Group C built on both of their efforts to do things that neither Group A nor Group B were capable of. This is what we do.

Outside that issue, another source of the confusion is that even if people have some understanding of how research works in general, they may not understand drug research in particular. This has come up over and over again here, but (put shortly) identifying a potential target, or coming up with a hypothesis about a disease, is not discovering a drug. To use the example above, Group B did not discover a drug. Many people seem to think that the steps from what those folks did to having a pill on the pharmacy shelves are rather routine and even trivial, but again: that’s what I’ve been working on since 1989, and believe me, neither of those adjectives apply. Group C, for example, might have found nothing in their screen, or they might very well manage to eventually take a compound into human patients and find (surprise!) that Group B’s proposal doesn’t work out as a therapy, or that such compounds also do other things that make them unusable. Happens about 90% of the time, actually – and I mean that completely literally: ninety per cent of the time things fail in the clinic. Just getting to patients at all from the Group B proposal level is no small achievement itself, and all that does is set you up for the chopping block I’ve just described.

So this post will be what I refer people to when the whole “Where Do Drugs Really Come From” issue comes up again, as it surely will. I have included my own self-serving Big Pharma Shill perspectives, I have included the views of the former dean of Harvard’s Medical School, those of an academic team at Bentley, those of a professor of intellectual property at Stanford, and those of the Congressional Research Service. As far as I can see, we’re pretty much on the same page. Anyone who has a different take on the subject will do themselves and the rest of the world a favor if they will first sit down and read these to see if their views hold up. On the other hand, if you have a different take on the subject but don’t want to engage with these other sources, please do the rest of the world a favor by doing something else with your time entirely.

43 comments on “Where Drugs Come From: A Comprehensive Look”

  1. Biodeals says:

    There were quite a few groups in industry that discovered and validated new biology and new targets – not sure if that has changed in big pharma (I am in BD now, so not as close to the lab as when I was discovering and validated targets). Discovery biologists in the industry were not always permitted to publish their findings until the associated compounds were patented, which is why they didn’t get credit in many cases (any publications happened years later). Not sure there is a way to track all of these contributions.

  2. Old Timer says:

    ““>90% of this funding represents basic research related to the biological targets for drug action rather than the drugs themselves“
    I don’t want to be pedantic, but I’m sure most of those name reactions/reagents that the chemists are using to do med. chem. and actually make the drug on scale were discovered under NIH funding–just a long time ago. So I would take issue with >90% biology, <10% chemistry.

    1. LD says:

      Hmm.. Let me see… Did you write your PhD, MA, whatever thesis in a Word created by Microsoft? 90% of your thesis then created by Microsoft based on your logic.

    2. UsedToBeAtIRBM says:

      And I don’t want to be pedantic, but I doubt most of the named reactions were funded by the NIH. Grignard, Suzuki, Arbuzov, Arndt-Eistert, Biginelli, Barton, Baeyer Villiger…. But then I’m not American so maybe I’m wrong.

      1. Old Timer says:

        I’m glad to hear that your libraries could have been made almost entirely before WWI. Pretty narrow list.
        I would also argue that not all biology takes place in the US, so the 90/10 split is even more suspect.

  3. Mark says:

    Here is a personal experience: recruiters and resume writers who have no idea how science works look at my (or Derek’s) work experience and career accomplishments with no understanding of what creates value.

    Their reply is (using Derek’s narrative above): “So you worked thirty years and you only developed how many drugs? Is that all you can did?”

    I’ve heard this more than once. How do you deal with people like that?

    1. Nick K says:

      It would take all my self-control not to reply with a brisk smack in the mouth.

  4. Bacillus says:

    I don’t know why Pharma companies don’t just make videos showing all the steps that go into getting a well known drug or vaccine to market and put them on YouTube and the likes. I know I was blown away when I visited GSKs facilities in Rixensart several years ago. Apparently, Cervarix has to undergo 300 different lot release assays which take 100 days to complete!

    1. nonchemist says:

      An alternative would be a reality show a la Survivor/The Bachelor, with several competing academic and industrial groups given a modest budget and asked to synthesize a series of drug candidates (against, say, an orphan disease instead of any of the big hitters).

      Their efforts are then televised each week, so that we can taste their mounting despair as compound after compound falls through the assay gauntlet.

      1. SV says:

        And the prize is priority FDA voucher!

      2. eub says:

        If you’re willing to accept drug discovery being dramatized in terms of burning adulterous glances plus colored lights on jungles of glassware, I think we can roll with this.

      3. David says:

        I would watch this.

        1. loupgarous says:

          A TV series based on “Things I Won’t Work With” would be much more entertaining. With Derek narrating in the background about “limb -to-chemist ratio” and “Satan’s kimchi” and Adam Savage setting up demonstrations for each episode.

          Dude, you need to find a screenwriter!

    2. C Wilson says:

      You are spot on about industry not helping itself very much. That said, Pfizer did recently support some commercials telling the tale of the arduous process, including failures (which we all pay for, but which are learnings as well). In addition there are a few books, one of which is John la Mattina’s DRUG TRUTHS. I used to work serving the commercial side of Pharma and recently moved to focus on the discovery and early development side. What a difference and what an appreciation I have developed for the complexities of biology and the discovery process. Heartfelt appreciation.

  5. Kelvin says:

    Here’s another thing to consider: Pharma only significantly profits from a drug it has developed for about 10 years while the patent lasts. After that, the drug invention and all its value belongs to society forever, so 99.99999999999999% of the value goes directly to society, while the rest goes back to society via the investors.

    In other words, people should stop being so narrow-minded in the way they think about these things.

    1. Anonymous says:

      W/o having read everything yet, a quick comment on Flier’s Fig 1, yellow balloon under Pharma/Biotech: “Develop chemical matter/potential drug”. I know that a summary figure is just that, a summary (I’ll read more later) but many lead compounds / chemical matter / potential drugs DO come out of NIH funded academic labs and research! That balloon should stretch over the Academia column.

      Re: Kelvin and comments on value. “Value” needs clarification each time it is used in this context. Dollar value, direct health/medical value, “social” value consequent to overall improved health, etc.

      People are concerned that much of the DOLLAR value accrues to the private sector and their often $multi-million compensated officers. (The Epipen story is a good example of that: “Proxy filings show that from 2007 to 2015, Mylan CEO Heather Bresch’s total compensation went from $2,453,456 to $18,931,068, a 671 percent increase. During the same period, the company raised EpiPen prices, with the average wholesale price going from $56.64 to $317.82, a 461 percent increase, according to data provided by Connecture.”)

      Once off-patent, the private sector still makes profits from many drugs, sometimes HUGE profits (see many previous stories In The Pipeline), and SOCIETY is usually PAYING, not reaping the $ benefits of those profits. Whether it is via government programs (Medicare, etc.) or private insurance premiums (paid by workers across ALL sectors), the public wants to know why the cost of essential, life saving generics have gone up so much. The appearance is: to make the fattest cats fatter.

      The last argument from Kelvin sounds like a variation of the trickle-down theory. Heather Bresch is using her $18 million salary to buy a car from Detroit and a computer from China and a cell phone from Korea or paying a house cleaner from a nearby town? Or she is putting some of it into a VC fund that will help a startup company or to make a new Hollywood movie?

      Personally, I think back to George Merck: “We try to remember that medicine is for the patient. We try never to forget that medicine is for the people. It is not for the profits. The profits follow, and if we have remembered that, they have never failed to appear. The better we have remembered it, the larger they have been.” George Merck is dead. In contemporary Pharma, so, it seems, is that ideal.

      1. Anon says:

        Even the dollar portion of value also gets cycled back to society as you can’t take the money with you to the grave. Through taxes, and paying others directly for work and/or consumables, it all goes back into the big pot eventually.

        1. Anon says:

          And even if you do try to hang on to the money, it’s just paper, and increases the value of money elsewhere by reducing overall supply of total money in effective circulation.

      2. StumpedByTheCaptchaMath says:

        Good example, but you picked the wrong Mylan executive to highlight. Bresch was the fallperson, Robert Coury was the mastermind behind it all. He was the CEO before her and moved into the Executive Chairman position calling all the shots while they groomed Bresch to take over. If you think Bresch’s $18.9M salary was high, hold onto the edge of your seat for Coury’s compensation…a whopping $97.6M in 2017 before quietly slipping away while Bresch was being hauled in front of congress for a public shaming.

  6. Jonathan says:

    Pharma only significantly profits from a drug it has developed for about 10 years while the patent lasts. After that, the drug invention and all its value belongs to society forever, so 99.99999999999999% of the value goes directly to society, while the rest goes back to society via the investors.

    If only that were still true. Insulin, adrenalin, and daraprim (to name three) were all discovered decades ago, but that hasn’t stopped some unscrupulous companies from benefiting massively from them recently, at direct cost to society.

    It’s a separate issue to the one of “who actually discovers the drugs”, but it’s one that gets easily conflated by non-experts, as we saw recently in the House of Representatives. Joe and Jane Sixpack don’t care when we first purified insulin (for example), they just care that they can’t afford lifesaving medicine because some evil cretin decided to jack the price up several hundred percent. And their actions tarnish the rest of the field.

  7. Geo says:

    Not so easy to get NIH funding for a drug discovery project within the University setting. If your work is highly novel so that it generates a new drug that will compete with and possibly replace a currently available commercial drug, you can expect to be judged by study section members who may be competitors and who participate on big pharma scientific boards or who are pharma consultants.

    1. MikeC says:

      Doubtful, considering that the patents on any current drugs will have expired before any new candidate would make it through testing. I’d worry more about the idea getting “borrowed”.

  8. John Wayne says:

    Expensive generics appear to be an unfortunate side effect of regulations designed for safety. There are too few people who are actually able to manufacture drug substances, creating an oligopoly of suppliers who have slowly (and sometimes not slowly) increased prices.

    If you think you can make something cheaper than the big boys, go ahead and get your GMP facility set up, product made, FDA approval, and sold to a formulation company. It’s a crazy task. It is far safer to invest in the market, start a fashion line, or toss some VC money at folks making freemium games for your phone.

    It is ironic that at the same time, the overall quality of the manufacturing for generics appears to have gone down. There have been a huge number of drug recalls in recent years. Most recently, the CVS brand recalled several medications for children.

  9. Biobot says:

    Drugs come from plants and other organisms. There are no exceptions. Who has the infrastructure to purify, chemical tweaks for optimizing, patent, go through the regulatory loops,…..that is another question.

    1. Industry Guy says:

      You should lay off the plants….

    2. Derek Lowe says:

      Lithium carbonate.

    3. Russian Bot says:

      Damn, this one got here first …

    4. SP123 says:

      Most synthetic chemistry reagents are ultimately from petroleum feed stocks, but I’m guessing that’s not the argument being made here…

    5. loupgarous says:

      Nitroglycerin, PEG, MRI contrast agents, the fentanyls, tramadol, the benzodiazepines, and (I think) all the monoamine reuptake inhibitors.
      Recent psychopharmacology seems to lean heavily on non-natural compounds.
      Pain control will probably do so in the future as new targets are identified which aren’t currently-known opioid receptors.

      1. Biobot says:

        No, I meant it mostly literally, if you look at the history of drugs, all of the major classes were pioneered by natural products and then tweaked by attorneys and med chemists to commercialize it. One of your examples–Tramadol is a good example of a “synthetic” drug that was later found to mysteriously have been independently synthesized by a tree. Actually Lowe did a good job of highlighting that finding. Coincidence… yeah I bet. The lithium is an interesting apparent counterexample but it seams to be ubiquitous in nature, including common minerals that people have consumed in water throughout history. Unfortunately, it was apparently pattented and so became a “drug”.

        1. loupgarous says:

          Good points, but “there are no exceptions” is sort of ignoring that an awful lot of drug synthesis was organic chemists making things and doctors like Paul Ehrlich (the Prussian one, not the American) trying things that had never been near a living organism until Ehrlich, et al made it druggable and tried it out on patients.

          One great big, huge exception back in Ehrlich’s time was arsenicals and mercury compounds, bequests to pharmacology from alchemy and (perhaps) mithridatism. At least, I’m not aware that anyone got their active ingredients from anywhere but a mine. I might be wrong, but I don’t think so.

          It’s only in the late 20th-early 21st century that we became aware that some animals require arsenic as a trace nutrient, and that things like New Caledonian sea sponges synthesize very interesting organoarsenic compounds like arsenicin A, which have an adamantane-like structure.

          1. loupgarous says:

            Whoops. Arsenic as an essential nutrient for rats, hamsters, goats, chickens, and maybe other animals is a 20th century thing. We were visiting a family friend who raised chickens and had a bottle of a veterinary arsenical to give to her chickens (for coccidiosis) which made my young hair stand on end.

        2. Derek Lowe says:

          No one could deny the contributions of natural products to drug discovery. But it’s also bizarre to deny the contributions of completely synthetic molecules (and leads). “Tweaked by attorneys and med chemists” is just not the reality. Have a look at the Njarson “200 drugs” posters and you’ll see that many of them have no natural product antecedents.

          You might also want to go back and read the part in the tramadol posts about the evidence that it isn’t a natural product at all, but contamination by the man-made compound.

          1. Nick K says:

            That poster of the 200 bestselling drugs is a work of art! How do I get a copy for my lab wall?

        3. Adonis says:

          Tramadol ain’t natural product. Would not hurt to read the full story.

  10. Industry R&D scientist says:

    OK, let’s accept all of these arguments (as I do, more or less).

    What do we do about the ~9% of the time that FDA approved drugs come directly from the publicly funded research? Take Lyrica, made by Rich Silverman and a postdoc at Northwestern with government funding. Under the current Bayh-Dole framework, Silverman and co. and Northwestern make hundreds of millions of the license and Pfizer makes tens of billions. Shouldn’t the taxpayers get a cut that can be reinvested into basic research?

    Or look at all the money sloshing around CRISPR right now. There are so many patents and licenses that come from Doudna, Zheng, and Church and are making “non-profits” like Berkeley/Broad/MIT/Harvard buckets. Should they be the only ones that get a cut?

    1. anoni says:

      I agree with your sentiment, but how many researchers at Berkeley/Broad/MIT/Harvard are not generating buckets of cash for their institutes? Their research, to some extent, is being aided by these income sources. So there is a bit of a loop here.

      1. Industry R&D scientist says:

        I was a PhD at one of those institutions and I can tell you that the departments and universities almost NEVER return the money from lucrative licensing deals back to specific research projects.

        More indirect funding, like operational costs, also typically come from grant overhead.

        Clearly the main goal of basic research is to benefit the public domain, so on average you won’t make money – but why is the approach for those uncommon but massively lucrative cases “socialize the risk but privatize the profit”?

        Alternatively, we could just make everything funded by a grant automatically public domain, but does anyone here want that?

        1. Anonymous says:

          Patent / royalty policies vary from institution to institution, but it is very roughly:
          one third to the inventors on the patent (in sometimes contested ratios)
          one third to the department or school
          one third to the university (sometimes directly to the patent / licensing office; sometimes to the general fund; Pres’s discretionary fund; etc.)
          AFTER subtraction of admin fees and other (patent filing, etc.) costs and anything else that can be skimmed off the top.

          Search “[university name] patent royalty policy” et sim to compare a few. (I just checked a few. The 3 x 1/3 rule of thumb is getting a lot more complicated, but I think it’s still a useful Rule of Thumb. Maybe it’s trending towards giving MORE to the inventors and their labs, up to 60:20:20)

          CU has an easy to read pie chart (link in my handle).

    2. Nameless says:

      You can try to come up with a special case that regulates these (very rare) cases and give the profits back to society. The other solution is to implement an effective corporate tax system (either with new laws or enforcing existing laws) and accept that some people will become disproportionally rich. The profits will work as a carrot to lure brilliant but naive (read young) people into the field and increase the overall success.
      At some point we just have to accept that no fair solution will work so concentrate on achieveable compromises.

  11. anon says:

    My impression is that there is skewed risk/reward balance between academia and industry. Most of the early researches are based on hypothesis and most of hypothesis are wrong (>90%). There are high risks in early research but not high reward. On the contrary, most people in industry are involved in processes that are not hypothesis driven and failure rates are much lower. Most processes pepole are paid with similar or even higher level of compensation than research people. This incentive scheme does not encourage people to do research. The current trend is the separation between research and process. Research people can sell their hypotheses, outsource the process and reap big reward if their hypotheis is correct ( If not, they are out of job). Process people will form CRO like entities that provide service and have stable income but less upside.

  12. CR says:

    “On the contrary, most people in industry are involved in processes that are not hypothesis driven and failure rates are much lower.”

    I doubt anyone on here would agree that failure rates in the drug industry and much lower.

  13. Ted Schroeder says:

    The most compelling analogue for the NIH, Academia and the Pharma/Biotech Industry collaboration is NASA, Academia and Industry. From computers, to cell phones, to 360 degree cameras, to memory foam, to GPS etc. the technologies that are so ubiquitous in making our lives so much easier were born from public/private partnerships. Sending a man to the moon and thereby expanding our understanding of the universe was the goal, enhancing the lives of earth bound humans was the benefit. Similarly, expanding our understanding of life through basic research is the goal, discovering life-saving and life-altering therapies is the benefit reaped by society. In between those two ideals industries from pharmaceuticals to chip makers, trillions of dollars have been generated which has inured (and continues to do so) an almost incalculable benefit to humankind. Not, just in technological benefit, but also in global living standards.

Comments are closed.